Aerodynamic parameter detection system and method for wind turbine blade

Abstract

The invention relates to the technical field of parameter detection, in particular to an aerodynamic parameter detection system and method for a wind turbine blade, and the system comprises a detection layout unit which is used for obtaining the structural design information of a target wind turbine blade, constructing a blade aerodynamic topological structure according to the structural design information, and carrying out the aerodynamic characteristic mutation analysis of the aerodynamic topological structure, determining an initial sensor layout area; the method comprises the following steps: collecting blade operation state data, performing time-frequency domain analysis on the operation state data, extracting airflow field characteristics and structure response characteristics, and generating a pneumatic-structure coupling sensitive descriptor; and the detection configuration unit is used for calculating the pneumatic sensitive index of each region of the blade based on the initial sensor layout region and the pneumatic-structure coupling sensitive descriptor. According to the method, the aerodynamic sensitive areas are determined by analyzing the aerodynamic characteristics of the blade, so that the sensors can be arranged in the areas, and the optimized layout can effectively improve the precision and efficiency of data acquisition and reduce the use of redundant sensors.

Description

Technical Field

[0001] This invention relates to the field of parameter detection technology, specifically to a system and method for detecting aerodynamic parameters of wind turbine blades. Background Art

[0002] With the increasing global demand for renewable energy, wind power is receiving more and more attention as a clean energy source. As a crucial component of the wind power system, the aerodynamic performance of wind turbine blades directly affects the efficiency and stability of the wind turbine. Therefore, real-time and accurate monitoring of the aerodynamic characteristics of the blades is essential for improving the operating efficiency and ensuring the safety of the wind turbine.

[0003] Currently, the deployment of sensors in traditional systems often lacks specificity. Sensors may be over-placed in areas where aerodynamic changes are not obvious, leading to wasted resources and data redundancy. In addition, such a layout cannot maximize monitoring accuracy and may fail to capture areas where aerodynamic characteristics change most significantly. Furthermore, there is no detailed analysis of the coupling relationship between aerodynamics and structure, and there is a lack of corresponding aerodynamic-structure coupling sensitive descriptors. Therefore, traditional systems may not be able to accurately grasp the complex interaction between airflow and blade structure, and cannot provide effective data support for performance optimization.

[0004] Furthermore, traditional systems often rely on static models or manual adjustments for aerodynamic parameter correction, unlike this invention which adjusts correction coefficients in real time based on actual operating conditions. This can lead to reduced monitoring accuracy when faced with blade aging or changes in the external environment, making it unable to effectively address the errors caused by these changes. Moreover, traditional systems often cannot perform synchronous sampling according to the airflow pulsation cycle, resulting in low utilization efficiency of sensor resources and a tendency to miss critical aerodynamic data. Consequently, the real-time performance and responsiveness of data acquisition are poor, and the data may not accurately reflect the actual operating conditions. Summary of the Invention

[0005] To achieve the above objectives, the present invention provides the following technical solution: an aerodynamic parameter detection system for wind turbine blades, comprising: The detection layout unit is used to acquire the structural design information of the target wind turbine blade, construct the blade aerodynamic topology based on the structural design information, perform aerodynamic characteristic mutation analysis on the aerodynamic topology, determine the initial sensor deployment area, collect blade operating status data, perform time-frequency domain analysis on the operating status data, extract airflow field characteristics and structural response features, and generate an aerodynamic-structure coupling sensitive descriptor. The detection configuration unit is used to calculate the aerodynamic sensitivity index of each region of the blade based on the initial sensor deployment area and the aerodynamic-structure coupling sensitive descriptor, optimize the sensor deployment position according to the aerodynamic sensitivity index, and generate a sensitive area detection configuration table. The calibration modeling unit is used to construct a blade aerodynamic monitoring simulation model based on the sensitive area detection configuration table, collect real-time aerodynamic data under different operating conditions through the aerodynamic monitoring simulation model, and construct an aerodynamic parameter adaptive calibration model based on the blade design aerodynamic parameters. The parameter correction unit is used to perform deviation analysis on real-time aerodynamic data using the aerodynamic parameter adaptive correction model, calculate aerodynamic parameter correction coefficients, and generate a dynamic detection execution plan based on the correction coefficients and the sensitive area detection configuration table. The aerodynamic detection unit is used to perform aerodynamic cycle sensing and scheduling of detection resources according to the dynamic detection execution scheme, establish a sampling sequence synchronized with the airflow pulsation cycle, and perform blade aerodynamic parameter detection.

[0006] Preferably, an aerodynamic topology of the blade is constructed based on the structural design information, and abrupt abrupt aerodynamic characteristic analysis is performed on the aerodynamic topology to determine the initial sensor deployment area, including: Obtain structural design information, including blade length, airfoil section parameters, chord length distribution, twist angle variation, and blade material properties; Based on the structural design information, a three-dimensional aerodynamic topology of the blade is constructed, and the surface mesh of the three-dimensional aerodynamic topology of the blade is divided and key sections are marked. Based on the computational fluid dynamics model, aerodynamic simulation of the topology is performed to analyze the airflow velocity, pressure distribution and aerodynamic load variation range of different cross sections and surface regions, and obtain aerodynamic parameter variation data. The locations of aerodynamic characteristic abrupt changes are identified based on aerodynamic parameter change data. These locations include areas where the airfoil curvature change exceeds a preset threshold, critical cross sections at the leading and trailing edges, extreme aerodynamic load locations, airflow separation risk areas, and pressure gradient abrupt change areas. The areas where these abrupt change locations are located are then determined as the initial sensor deployment areas.

[0007] Preferably, the operating status data is analyzed in the time and frequency domain to extract airflow field characteristics and structural response features, generating an aerodynamic-structural coupling sensitive descriptor, including: Time-frequency domain decomposition was performed on the pressure pulsation signal and vibration acceleration signal in the operating status data to obtain airflow separation characteristics and flutter mode characteristics; The turbulence intensity and pressure gradient change rate corresponding to the airflow separation characteristics, as well as the vibration amplitude and frequency characteristics corresponding to the flutter mode characteristics, are extracted to obtain the structural response characteristics; The correlation between airflow field characteristics and structural response features is analyzed, an aerodynamic-structural coupling relationship set is constructed, and an aerodynamic-structural coupling sensitive descriptor is generated based on the airflow state identifier and structural response identifier mapped from the coupling relationship set.

[0008] Preferably, based on the initial sensor deployment area and the aerodynamic-structural coupling sensitive descriptor, the aerodynamic sensitivity index of each region of the blade is calculated, including: Extract airflow state identifiers and structural response identifiers from the aero-structure coupling sensitive descriptor; The basic sensitivity coefficient is obtained by mapping the airflow state indicator to the turbulence level, and the structural influence coefficient is obtained by analyzing the vibration intensity of the structural response indicator. The basic sensitivity coefficient and structural influence coefficient are weighted and fused according to preset weights to calculate the aerodynamic sensitivity index of each region of the blade.

[0009] Preferably, the sensor placement is optimized based on the aerodynamic sensitivity index, generating a sensitive area detection configuration table that includes detection priority and accuracy requirements, including: The initial sensor deployment areas are prioritized based on the aerodynamic sensitivity index to obtain a region priority sequence. Based on the aerodynamic monitoring accuracy requirements of the blades, corresponding detection accuracy requirements are matched for different priority areas, and a set of regional accuracy constraints is generated. The sensor deployment density and location are adjusted according to the regional priority sequence and regional accuracy constraint set to generate a sensitive area detection configuration table, wherein the sensitive area detection configuration table includes detection priority, accuracy requirements and sensor location.

[0010] Preferably, a blade aerodynamic monitoring simulation model is constructed based on the sensitive area detection configuration table, and real-time aerodynamic data under different operating conditions is collected through the aerodynamic monitoring simulation model, including: Pressure sensors, wind speed sensors, and angle-of-attack sensors are deployed according to the sensor locations in the sensitive area detection configuration table, and the sensor data is associated with the grid nodes of the aerodynamic topology. The interpolation density of the aerodynamic monitoring simulation model is set, and the sensor-collected data is calibrated as known sampling points based on the radial basis function interpolation algorithm. The surface arc length distance between the point to be interpolated and the known sampling points is calculated, and the interpolation weight is determined. Based on the radial basis function interpolation algorithm, the aerodynamic information of known sampling points is interpolated to the grid nodes to be interpolated, and the surface pressure, airflow velocity and airflow angle of attack data of each node are obtained. The aerodynamic data of each node is mapped to a visual color gradient to generate a distribution map of blade aerodynamic information. The distribution map is refreshed in real time according to the preset sensor data update frequency. The dynamic changes of aerodynamic parameters on the blade surface can be observed through the distribution map to achieve real-time aerodynamic monitoring. Simulate different operating conditions and collect surface pressure, airflow velocity, and airflow angle of attack data under each condition through an aerodynamic monitoring simulation model as real-time aerodynamic data.

[0011] Preferably, an adaptive aerodynamic parameter correction model is constructed based on the blade design aerodynamic parameters, including: Obtain the aerodynamic parameters of the blade design and the actual aerodynamic requirements during operation. Based on the conservation laws of fluid mechanics and the principle of load balance of the blade structure, analyze the balance relationship between the design aerodynamic parameters and the structural load. Based on the aforementioned balance relationship and the correlation rules in the aerodynamic-structure coupling sensitive descriptor, the aerodynamic-structure coupling relationship is determined; Using real-time aerodynamic data as input and designed aerodynamic parameters as target output, an adaptive correction model for aerodynamic parameters is constructed. The adaptive correction model for aerodynamic parameters includes a deviation calculation module and a correction coefficient generation module, which are used to output the aerodynamic parameter deviation values ​​and the corresponding correction coefficients.

[0012] Preferably, the aerodynamic parameter adaptive correction model is used to perform deviation analysis on real-time aerodynamic data, calculate aerodynamic parameter correction coefficients, and generate a dynamic detection execution plan based on the correction coefficients and the sensitive area detection configuration table, including: Real-time aerodynamic data is input into the aerodynamic parameter adaptive correction model to calculate the aerodynamic pressure deviation and airflow velocity deviation at each monitoring point. Based on the magnitude and distribution of the deviation values, and combined with the turbulence level and vibration intensity in the aerodynamic-structure coupling sensitive descriptor, the aerodynamic parameter correction coefficients are calculated. Based on the detection priority and accuracy requirements in the sensitive area detection configuration table, adjust the sampling frequency and data filtering intensity of different areas to determine the detection resource allocation scheme; By integrating the correction coefficients, sampling frequency, and resource allocation scheme, a dynamic detection execution scheme is generated.

[0013] Preferably, the detection resources are aerodynamically cycle-sensing and scheduled according to the dynamic detection execution scheme, a sampling sequence synchronized with the airflow pulsation cycle is established, and blade aerodynamic parameter detection is performed, including: The detection tasks in the dynamic detection execution plan are analyzed, and the sensor types, quantities, and working times required for each task are calculated to obtain the sensor resource requirement set. Assess the available resources of the detection equipment, generate a sensor resource distribution map, and extract the airflow pulsation period based on the distribution map; Sampling time slots are divided according to the airflow pulsation cycle, and a sampling timing scheme synchronized with the cycle is established; According to the sampling timing scheme, the control sensor collects aerodynamic data, and the collected data is uploaded in real time through the data transmission module to perform blade aerodynamic parameter detection.

[0014] A method for detecting aerodynamic parameters of a wind turbine blade, applicable to the aforementioned aerodynamic parameter detection system for a wind turbine blade, comprising: Obtain the structural design information of the target wind turbine blade, construct the blade aerodynamic topology based on the structural design information, perform aerodynamic characteristic mutation analysis on the aerodynamic topology, and determine the initial sensor deployment area; Collect blade operating status data, perform time-frequency domain analysis on the operating status data, extract airflow field characteristics and structural response features, and generate aerodynamic-structural coupling sensitive descriptors; Based on the initial sensor deployment area and the aerodynamic-structure coupling sensitive descriptor, the aerodynamic sensitivity index of each region of the blade is calculated. The sensor deployment position is optimized according to the aerodynamic sensitivity index, and a sensitive area detection configuration table including detection priority and accuracy requirements is generated. A blade aerodynamic monitoring simulation model is constructed based on the sensitive area detection configuration table. Real-time aerodynamic data under different operating conditions are collected through the aerodynamic monitoring simulation model. An aerodynamic parameter adaptive correction model is constructed based on the blade design aerodynamic parameters. The aerodynamic parameter adaptive correction model is used to perform deviation analysis on real-time aerodynamic data, calculate aerodynamic parameter correction coefficients, and generate a dynamic detection execution plan based on the correction coefficients and the sensitive area detection configuration table. According to the dynamic detection execution scheme, the detection resources are aerodynamically cycle-sensing and scheduling is performed to establish a sampling sequence synchronized with the airflow pulsation cycle, and the blade aerodynamic parameters are detected.

[0015] Compared with the prior art, the present invention has the following beneficial effects: (1) This invention analyzes the aerodynamic characteristics of the blades to determine the aerodynamic sensitive areas, thereby enabling the deployment of sensors in these areas. This optimized layout can effectively improve the accuracy and efficiency of data acquisition and reduce the use of redundant sensors. Through time-frequency domain analysis, the coupling relationship between airflow characteristics and blade structure response is extracted to generate an aerodynamic-structure coupling sensitive descriptor. Such a descriptor can help engineers understand the complex interaction between airflow and blade structure, providing data support for accurate monitoring and optimization of blade performance. (2) Through the aerodynamic parameter adaptive correction model, the system can adjust the correction coefficient in real time during actual operation, perform deviation analysis on aerodynamic parameters, and maintain high accuracy of monitoring data. This dynamic update mechanism enables the system to adapt to various working environments, reduce errors caused by blade aging or changes in external conditions, and can synchronously schedule the sampling sequence according to the airflow pulsation cycle to optimize the utilization rate of sensor resources and ensure the consistency between data acquisition and actual working conditions. Through this efficient scheduling, the system can accurately capture aerodynamic data at critical moments and improve the real-time performance and responsiveness of detection. (3) Through simulation based on real-time aerodynamic data and aerodynamic topology, the system can update the blade aerodynamic distribution map in real time and provide engineers with visualized aerodynamic information to help them understand the operating status of the blade and make optimization adjustments. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the overall system architecture in one embodiment of the present invention; Figure 2 This is a schematic flowchart of the overall method in one embodiment of the present invention.

[0017] In the diagram: 1. Detection layout unit; 2. Detection configuration unit; 3. Calibration modeling unit; 4. Parameter calibration unit; 5. Pneumatic detection unit. Detailed Implementation

[0018] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0019] Example 1, please refer to Figure 1 This invention provides a technical solution: an aerodynamic parameter detection system for wind turbine blades, comprising: The detection layout unit 1 is used to acquire the structural design information of the target wind turbine blade, construct the blade aerodynamic topology based on the structural design information, perform aerodynamic characteristic mutation analysis on the aerodynamic topology, determine the initial sensor deployment area, collect blade operating status data, perform time-frequency domain analysis on the operating status data, extract airflow field characteristics and structural response features, and generate an aerodynamic-structure coupling sensitive descriptor. The detection configuration unit 2 is used to calculate the aerodynamic sensitivity index of each region of the blade based on the initial sensor deployment area and the aerodynamic-structure coupling sensitive descriptor, optimize the sensor deployment position according to the aerodynamic sensitivity index, and generate a sensitive area detection configuration table. The calibration modeling unit 3 is used to build a blade aerodynamic monitoring simulation model based on the sensitive area detection configuration table, collect real-time aerodynamic data under different operating conditions through the aerodynamic monitoring simulation model, and build an aerodynamic parameter adaptive calibration model based on the blade design aerodynamic parameters. The parameter correction unit 4 is used to perform deviation analysis on real-time aerodynamic data using an aerodynamic parameter adaptive correction model, calculate aerodynamic parameter correction coefficients, and generate a dynamic detection execution plan based on the correction coefficients and the sensitive area detection configuration table. The aerodynamic detection unit 5 is used to perform aerodynamic cycle sensing and scheduling of detection resources according to the dynamic detection execution plan, establish a sampling sequence synchronized with the airflow pulsation cycle, and perform blade aerodynamic parameter detection.

[0020] It should be noted that the process involves acquiring the design information (such as size, shape, and material) of the target wind turbine blade, constructing an aerodynamic topology model of the blade based on this information, and performing aerodynamic characteristic mutation analysis. The analysis results are used to determine the initial sensor deployment area. For example, assuming a new type of wind turbine blade is designed with a long leading edge and a special curvature, the detection deployment unit will construct an aerodynamic topology model based on this design data to identify which parts of the blade are likely to be significantly affected by wind speed changes (such as the leading edge region and blade tip). These areas will be selected as the initial sensor deployment area. During wind turbine operation, the system collects actual operating state data of the blade. By performing time-frequency domain analysis on this data, airflow can be extracted. The field characteristics and blade structure response features are used to generate aerodynamic-structural coupling sensitivity descriptors. For example, assuming the wind turbine operates at different wind speeds, sensors record blade vibration and pressure data. Time-frequency domain analysis may reveal that the airflow at the blade tip is highly unstable and generates significant vibration, indicating that this region is very sensitive to airflow changes. This analysis result generates the aerodynamic-structural coupling sensitivity descriptor for subsequent optimization. Based on the initially determined sensor deployment area and the aerodynamic-structural coupling sensitivity descriptor, the aerodynamic sensitivity index for each region is calculated, and the sensor deployment positions are optimized. Finally, a sensitive area detection configuration table is generated. For example, based on the previous analysis, assuming the blade tip region is highly sensitive to wind... The blade is most sensitive to speed changes. The detection configuration unit calculates the aerodynamic sensitivity index for each region based on the sensitivity descriptor, determining that the sensitivity index is higher in the tip region. Therefore, it optimizes the sensor layout by increasing the number of sensors in that region. Based on the optimized sensor layout, an aerodynamic monitoring simulation model of the blade is constructed. Through this model, the system can collect real-time aerodynamic data under different operating conditions and adjust the model according to the blade's design aerodynamic parameters, thereby performing adaptive aerodynamic parameter correction. For example, assuming that the wind turbine blade performs differently than expected under different wind speed conditions (such as excessively high or low wind speeds), the correction modeling unit adjusts the aerodynamic parameter model based on the real-time collected data and the blade's design aerodynamic parameters to more accurately reflect the actual operating conditions. The system analyzes the aerodynamic performance of the wind turbine blades under various conditions. For example, if excessive wind speed causes abnormal aerodynamic parameters in a certain area, the model will automatically adjust the aerodynamic parameters of that area based on real-time data. It uses an adaptive correction model to perform deviation analysis on real-time aerodynamic data and calculates correction coefficients for the aerodynamic parameters. Based on the correction coefficients and a sensitive area detection configuration table, a dynamic detection execution plan is generated. For example, if the collected data reveals a difference between the aerodynamic load on the wind turbine blades and the design expectation, the parameter correction unit will analyze these deviations using an adaptive correction model and calculate correction coefficients. For instance, if the aerodynamic pressure deviation in certain areas is too large, the system will calculate an adjustment factor and generate a new dynamic detection execution plan to adjust these deviations.Function: Based on the dynamic detection execution plan, the system performs real-time aerodynamic cycle sensing and scheduling, and synchronizes detection with the airflow pulsation cycle to ensure that the sampling sequence is consistent with the airflow cycle, thereby improving detection accuracy. For example, during wind turbine operation, airflow changes are periodic; the aerodynamic detection unit senses these airflow pulsation cycles in real time and synchronizes the sampling sequence with the airflow fluctuation cycle. For instance, when the airflow enters a fluctuation phase, the system will collect more data, thereby improving detection accuracy and ensuring that changes in the aerodynamic characteristics of the blades are captured.

[0021] In an optional embodiment, an aerodynamic topology of the blade is constructed based on structural design information, and abrupt abrupt aerodynamic characteristic analysis is performed on the aerodynamic topology to determine the initial sensor deployment area, including: Obtain structural design information, including blade length, airfoil section parameters, chord length distribution, twist angle variation, and blade material properties; Based on the structural design information, a three-dimensional aerodynamic topology of the blade is constructed, and the surface mesh of the three-dimensional aerodynamic topology of the blade is divided and key sections are marked. Based on the computational fluid dynamics model, aerodynamic simulation of the topology is performed to analyze the airflow velocity, pressure distribution and aerodynamic load variation range of different cross sections and surface regions, and obtain aerodynamic parameter variation data. The locations of aerodynamic characteristic abrupt changes are identified based on aerodynamic parameter change data. These locations include areas where the airfoil curvature change exceeds a preset threshold, key sections at the leading and trailing edges, extreme aerodynamic load locations, airflow separation risk areas, and pressure gradient abrupt change areas. The areas where these abrupt changes occur are then designated as the initial sensor deployment areas.

[0022] It should be noted that the system needs to obtain the structural design information of the wind turbine blades. This design information includes: blade length: the total length of the blade, which directly affects the wind-catching area of ​​the wind turbine; airfoil section parameters: the cross-sectional shape of the blade at each location, which determines the interaction between the airflow and the blade; chord length distribution: the variation in blade width, which typically decreases gradually from the blade root to the tip; twist angle variation: the degree of twist of the blade at different locations, affecting the aerodynamic efficiency of the blade; blade material properties: the strength, stiffness, weight, etc., of the materials used in the blade, affecting its rigidity and durability. For example, suppose a new type of wind turbine blade is designed with a long leading edge and a certain degree of curvature; the system will record these data when obtaining the design information. For example, leading edge length, airfoil (such as NACA4412), and chord length (wider at the blade root and narrower at the tip) are used to construct the blade's aerodynamic model. The system establishes a three-dimensional aerodynamic topology of the blade based on the structural design information and meshes the blade surface for aerodynamic analysis. It also marks critical sections on the blade. Specifically, based on the blade's design data, the system creates a three-dimensional aerodynamic topology model. Assuming the blade design has a large curvature region (such as a curved leading edge), where airflow changes significantly, the system marks these regions as critical sections, which are typically high-risk areas for abrupt aerodynamic changes. Computational fluid dynamics (CFD) is then used to construct the aerodynamic model. The model performs aerodynamic simulations of the three-dimensional topology of the blade. The simulation displays changes in airflow velocity, pressure distribution, and aerodynamic loads in different regions. Ultimately, the system obtains data on aerodynamic parameter variations to analyze the blade's aerodynamic performance. For example, assuming a wind turbine blade has a long curvature at its leading edge, the CFD simulation reveals that the airflow velocity at the blade's leading edge changes drastically at certain wind speeds, causing fluctuations in local pressure and aerodynamic loads. These regions may exhibit abrupt changes in aerodynamic characteristics, requiring special attention. The system identifies the locations of these abrupt changes based on the aerodynamic parameter variation data. These abrupt changes typically include areas where the airfoil curvature changes beyond a preset threshold; for example, the leading edge region of the blade may experience abrupt changes. Significant curvature changes can affect airflow; critical sections at the leading and trailing edges, typically near the blade root and tip, are areas where wind speed and airflow variations are drastic; extreme aerodynamic load locations are areas where wind speed changes can generate significant aerodynamic loads; airflow separation risk areas are regions where airflow separation can occur, causing the blade to lose effective aerodynamic performance; abrupt pressure gradient changes are areas where airflow pressure changes drastically, which are also areas where aerodynamic performance changes; for example, during simulation, if the leading edge curvature of a wind turbine blade changes significantly, the system will detect this change and determine that airflow separation is likely to occur in this area, thus marking this area as an aerodynamic characteristic abrupt change area.The system may also detect abrupt changes in pressure gradient at the blade tip, an area requiring special attention. Based on the location of these aerodynamic abrupt changes, the system will determine the initial sensor deployment area. These areas are typically those with severe aerodynamic abrupt changes or high sensitivity to wind speed variations. Sensors will be deployed in these areas to monitor aerodynamic parameters in real time. For example, the leading edge and tip regions are identified as high-risk areas for aerodynamic abrupt changes, and sensors will be deployed in these areas. If the curvature change in the leading edge region exceeds a preset threshold, this area is highly sensitive to airflow changes, and therefore multiple sensors will be placed at this location to monitor aerodynamic performance.

[0023] In an optional embodiment, time-frequency domain analysis is performed on the operating status data to extract airflow field characteristics and structural response features, generating an aerodynamic-structural coupling sensitive descriptor, including: Time-frequency domain decomposition was performed on the pressure pulsation signal and vibration acceleration signal in the operating status data to obtain airflow separation characteristics and flutter mode characteristics; The turbulence intensity and pressure gradient change rate corresponding to the airflow separation characteristics, as well as the vibration amplitude and frequency characteristics corresponding to the flutter mode characteristics, are extracted to obtain the structural response characteristics; The correlation between airflow field characteristics and structural response features is analyzed, an aerodynamic-structural coupling relationship set is constructed, and airflow state identifiers and structural response identifiers are mapped based on the coupling relationship set to generate an aerodynamic-structural coupling sensitive descriptor.

[0024] It should be noted that time-frequency domain analysis transforms pressure pulsation signals and vibration acceleration signals into easily understandable features, aiding in the identification of airflow separation and flutter mode characteristics. For example, suppose a pressure sensor is installed on the leading edge of a wind turbine blade, detecting periodic fluctuations in surface pressure at a certain moment. These fluctuations may be caused by airflow separation. Through time-frequency domain decomposition, the system identifies the frequency of these fluctuations, perhaps finding it to be around 10Hz, indicating that airflow separation occurs within this frequency range. On the other side of the blade, a vibration acceleration sensor records a vibration peak at 15Hz. By analyzing the time-frequency domain signal, the system can determine that this frequency is related to flutter modes. The system then extracts the data related to airflow separation and flutter modes. The characteristics, specifically, include turbulence intensity, pressure gradient change rate, vibration amplitude, and frequency characteristics. Airflow separation characteristics include: turbulence intensity: indicating the degree of airflow disorder; turbulence intensity is usually high at the point of separation; pressure gradient change rate: drastic pressure changes occur in the airflow separation region, leading to a large change in the pressure gradient; vibration amplitude: representing the vibration intensity of the blade at a specific frequency; flutter modes typically exhibit large vibration amplitudes at specific frequencies; frequency characteristics: different flutter modes have different frequencies. For example, the system detected a turbulence intensity of 0.65 in a certain region, indicating that the airflow in that region is very turbulent and airflow separation may have occurred; simultaneously, the pressure gradient change rate is 0.6 Pa / m, further confirming the airflow separation in that region. Dramatic changes were observed; for blade vibration, the system recorded a signal with an amplitude of 3 mm / s² at a frequency of 20 Hz, indicating severe flutter mode at this frequency. Based on airflow separation and flutter mode characteristics, structural response features were extracted to evaluate the wind turbine blade's response to airflow changes. Structural response features included: vibration response: the blade's vibration response at a specific frequency, typically related to airflow turbulence intensity and pressure changes; stress and deformation: the stress state and deformation of the wind turbine blade, especially in airflow separation or flutter regions; for example, in the region of wind turbine blade with a turbulence intensity of 0.65 and a pressure gradient change rate of 0.6 Pa / m, the system detected a stress value of 4.5 MPa. The deformation of 1.0 mm indicates significant stress concentration and deformation in the airflow separation region. Under the 20 Hz flutter mode, the vibration amplitude was 3 mm / s², causing the blade stress to reach 5 MPa in some areas, triggering a real-time alarm and indicating potential flutter risk. Analysis of the correlation between airflow field characteristics and structural response features revealed the interrelationship between airflow and blade response, providing a basis for further optimization and safety assessment. For example, the system found that when the turbulence intensity exceeded 0.6, the blade vibration amplitude reached its maximum at 20 Hz, and the blade stress also began to increase significantly. This indicates that the turbulence characteristics of the airflow directly affect the blade's vibration and stress state. In another region, the turbulence intensity was 0.5. Small variations in vibration amplitude and stress indicate that the relatively stable airflow has a minimal impact on the blades. Based on airflow state and structural response indicators, an aerodynamic-structural coupling relationship set is constructed, and sensitive descriptors are generated for real-time monitoring and prediction of the wind turbine's operating status (such as vibration amplitude and stress). For example, when the turbulence intensity reaches 0.6, the vibration amplitude is 3 mm / s², and the stress value is 4.5 MPa, the system generates a sensitive descriptor, identifying the area as a high-risk region. The system automatically issues a warning and suggests measures to mitigate the risk of airflow separation or flutter.

[0025] In an optional embodiment, based on the initial sensor deployment area and the aerodynamic-structural coupling sensitivity descriptor, the aerodynamic sensitivity index of each region of the blade is calculated, including: Extract airflow state identifiers and structural response identifiers from the aero-structure coupling sensitive descriptor; The basic sensitivity coefficient is obtained by mapping the airflow state indicator to the turbulence level, and the structural influence coefficient is obtained by analyzing the vibration intensity of the structural response indicator. The basic sensitivity coefficient and structural influence coefficient are weighted and fused according to preset weights to calculate the aerodynamic sensitivity index of each region of the blade.

[0026] It should be noted that airflow state information and structural response information are extracted from the already generated aerodynamic-structural coupled sensitivity descriptors. For example, suppose that in a certain region of the leading edge of a wind turbine blade, a turbulence intensity of 0.7 is obtained through time-frequency domain analysis, indicating that the airflow in this region is very unstable. Simultaneously, the blade vibration amplitude in this region is recorded as 2.5 mm / s², which corresponds to the structural response descriptor. By classifying the turbulence intensity in the airflow state descriptors and mapping it to a basic sensitivity coefficient, the degree of airflow influence is measured. Based on different turbulence intensities, they are divided into different levels (e.g., low, medium, high). (High turbulence), each level corresponds to a basic sensitivity coefficient; the basic sensitivity coefficient represents the degree of influence of airflow on wind turbine blades; the higher the turbulence intensity, the larger the basic sensitivity coefficient, indicating a greater influence of airflow on the blades; the structural influence coefficient is calculated by analyzing the structural response indicator (i.e., the vibration response of the blades); vibration amplitude is usually an important indicator of the wind turbine blades' response to changes in airflow; by analyzing the vibration amplitude, the severity of blade vibration is determined; the structural influence coefficient represents the degree of influence of structural response (such as blade vibration) on the overall performance and safety of the wind turbine; the larger the vibration amplitude, the higher the structural influence coefficient; for example: assuming that at a certain point on the blade... In the region, the recorded vibration amplitude was 2.5 mm / s². Based on preset vibration intensity levels: low vibration intensity (0-1.0 mm / s²) corresponds to a structural influence coefficient of 0.3, medium vibration intensity (1.0-3.0 mm / s²) corresponds to a structural influence coefficient of 0.6, and high vibration intensity (above 3.0 mm / s²) corresponds to a structural influence coefficient of 1.0. Therefore, the structural influence coefficient for the region with a vibration amplitude of 2.5 mm / s² is 0.6. The foundation sensitivity coefficient and structural influence coefficient are weighted and fused according to preset weights to comprehensively evaluate each region. The aerodynamic sensitivity is determined by combining the influence of airflow state on structural response (basic sensitivity coefficient) with the influence of structural response (structural influence coefficient), typically using preset weight values. These weights can be adjusted based on actual conditions. For example, assuming the preset weights are: airflow state weight 0.6, structural response weight 0.4; then, the airflow state indicator (basic sensitivity coefficient) for the leading edge region is 0.8, the structural response indicator (structural influence coefficient) is 0.6, and the weighted aerodynamic sensitivity index is: Aerodynamic sensitivity index = 0.6 × 0.8 + 0.4 × 0.6 = 0.48 + 0.24 = 0.72. By calculating the aerodynamic sensitivity index of each region of the blade using weighted fusion, the aerodynamic performance and potential risks of each region are obtained. The aerodynamic sensitivity index reflects the coupled influence of airflow and structural response, and is typically used to assess the aerodynamic performance of the blade and identify potential risk areas. For example, assuming a wind turbine blade has multiple regions, each region undergoes similar calculations, yielding different aerodynamic sensitivity indices. For instance, Region 1: aerodynamic sensitivity index 0.72 (as shown above), Region 2: aerodynamic sensitivity index 0.55, Region 3: aerodynamic sensitivity index 0.87. Based on these aerodynamic sensitivity indices, it can be determined which part of the blade is in a higher-risk state. For example, a higher aerodynamic sensitivity index in Region 3 indicates that this region may face higher aerodynamic loads and structural risks.

[0027] In an optional embodiment, the sensor placement is optimized based on the aerodynamic sensitivity index to generate a sensitive area detection configuration table that includes detection priority and accuracy requirements, including: The initial sensor deployment areas are prioritized based on the aerodynamic sensitivity index to obtain a region priority sequence. Based on the aerodynamic monitoring accuracy requirements of the blades, corresponding detection accuracy requirements are matched for different priority areas, and a set of regional accuracy constraints is generated. The sensor deployment density and location are adjusted based on the regional priority sequence and regional accuracy constraint set to generate a sensitive area detection configuration table, which includes detection priority, accuracy requirements and sensor location.

[0028] It should be noted that by calculating the aerodynamic sensitivity index of each region of the blade and ranking these regions according to their aerodynamic sensitivity, it is possible to determine which regions require more monitoring resources. The aerodynamic sensitivity index of each region is then ranked from highest to lowest. Regions with higher aerodynamic sensitivity indices indicate a greater coupling effect between airflow and structure, thus requiring more sensor monitoring. For example, assuming the calculated aerodynamic sensitivity indices for three regions of the blade are: Region 1: 0.72, Region 2: 0.55, Region 3: 0.87, then Region 3 has the highest aerodynamic sensitivity index. Area 3 has the highest priority and requires the fastest sensor deployment; Area 2 has the lowest aerodynamic sensitivity index, therefore its priority is the lowest; the priority sequence is: Area 3 > Area 1 > Area 2, with different monitoring accuracy requirements set for each area; higher priority areas typically require higher monitoring accuracy, while lower priority areas can have lower accuracy requirements; different areas require different monitoring accuracies due to their different aerodynamic characteristics and structural responses; generally, higher priority areas require higher monitoring accuracy to ensure the accuracy of data in critical areas; for example: Area 3 (highest priority, aerodynamic sensitivity index of 0.87): requires high-precision monitoring, with accuracy requirements... The accuracy requirement is ±0.1 for Region 1 (aerodynamic sensitivity index of 0.72), and ±0.2 for Region 2 (aerodynamic sensitivity index of 0.55). The generated region accuracy constraint sets are as follows: Region 3: accuracy requirement ±0.1, Region 1: accuracy requirement ±0.2, Region 2: accuracy requirement ±0.3. Based on the region priority and accuracy requirements, the sensor deployment density and location are adjusted reasonably to ensure higher density sensors in key areas to meet high accuracy requirements. Based on the region priority and accuracy requirements, an appropriate number of sensors and their deployment locations are selected for each region. Higher priority areas may require more sensors and a finer deployment density. For example, based on the previously generated area priority sequence and accuracy requirements, assuming sensors can be deployed on the blade surface, the following deployment strategy can be adopted: Area 3 (highest priority): requires more sensors, with a deployment density of 2 sensors per square meter, and the location should be selected in the area with the strongest airflow disturbance; Area 1: requires a medium density of sensors, with a deployment density of 1.5 sensors per square meter, and the location should be placed in the area where the aerodynamic performance of the blade is more sensitive; Area 2 (lowest priority): requires a deployment density of 1 sensor per square meter, and the location should be selected in the area with less airflow influence.

[0029] In an optional embodiment, a blade aerodynamic monitoring simulation model is constructed based on a sensitive area detection configuration table. Real-time aerodynamic data under different operating conditions is collected through this simulation model, including: Pressure sensors, wind speed sensors, and angle-of-attack sensors are deployed according to the sensor locations in the sensitive area detection configuration table, and the sensor data is associated with the grid nodes of the aerodynamic topology. The interpolation density of the aerodynamic monitoring simulation model is set, and the sensor-collected data is calibrated as known sampling points based on the radial basis function interpolation algorithm. The surface arc length distance between the point to be interpolated and the known sampling points is calculated, and the interpolation weight is determined. Based on the radial basis function interpolation algorithm, the aerodynamic information of known sampling points is interpolated to the grid nodes to be interpolated, and the surface pressure, airflow velocity and airflow angle of attack data of each node are obtained. The aerodynamic data of each node is mapped to a visual color gradient to generate a distribution map of blade aerodynamic information. The distribution map is refreshed in real time according to the preset sensor data update frequency. The dynamic changes of aerodynamic parameters on the blade surface can be observed through the distribution map to achieve real-time aerodynamic monitoring. Simulate different operating conditions and collect surface pressure, airflow velocity, and airflow angle of attack data under each condition through an aerodynamic monitoring simulation model as real-time aerodynamic data.

[0030] It should be noted that, based on the priority, accuracy requirements, and sensor deployment density in the sensitive area detection configuration table, various sensors are deployed on the blade surface, and the collected data is associated with the grid nodes of the aerodynamic topology to facilitate subsequent interpolation and analysis. According to the sensor deployment density and location in the sensitive area detection configuration table, pressure sensors, wind speed sensors, and angle-of-attack sensors are deployed to their respective areas. The data collected by each sensor (e.g., pressure, wind speed, angle of attack) will correspond one-to-one with the nodes of the aerodynamic topology grid. For example, based on the configuration table from the previous step, assuming the blade surface is divided into multiple grid nodes, the sensors will be deployed in the following locations: Area 3 (highest priority) will have pressure sensors, wind speed sensors, and angle-of-attack sensors deployed. Sensors are deployed in high density near areas with significant airflow disturbance; Region 1 (medium priority) will have a certain number of sensors deployed, primarily monitoring areas more sensitive to the aerodynamic performance of the blades; Region 2 (lowest priority) will have a lower density of sensors deployed, primarily monitoring areas less affected by airflow. The interpolation density of the aerodynamic monitoring simulation model is set, and sensor-collected data are calibrated as known sampling points based on the radial basis function interpolation algorithm. The surface arc distance between the point to be interpolated and the known sampling points is calculated to determine the interpolation weights. To more accurately simulate the aerodynamic characteristics of the entire blade surface, an interpolation algorithm is used to fill areas without sensors. By setting the interpolation density, sensor-collected data are calibrated as known sampling points, and the interpolation weights for each point to be interpolated are calculated. The interpolation points are weighted relative to known sampling points; the interpolation accuracy is set, i.e., the distance between the interpolation points in the simulated grid nodes; the radial basis function (RBF) algorithm is used for interpolation to transfer the data collected by the sensors to other unsampled grid nodes; for example, assuming the blade surface is divided into 100 grid nodes, and only 30 nodes are equipped with sensors; during the interpolation process, the radial basis function interpolation method is used to extrapolate the 30 collected data points (e.g., pressure, wind speed, angle of attack) to the other 70 unsampled nodes; the interpolation weight is determined by calculating the surface arc length distance between each interpolation point and the known data points; the collected data is extended to all grid nodes on the entire blade surface using the interpolation algorithm. The process involves several steps: First, complete aerodynamic data is obtained by interpolating the sensor-collected data (pressure, wind speed, angle of attack) to obtain the aerodynamic parameters of the unsampled grid nodes. This data can then be used for further analysis and visualization. For example, after interpolation, suppose there is a grid node without sensors, which originally only had pressure and wind speed data (obtained through sensors). Through the interpolation algorithm, the airflow angle of attack value of this node will be calculated. Ultimately, all 100 grid nodes will have complete aerodynamic data (including surface pressure, wind speed, angle of attack, etc.). The aerodynamic data of each node is mapped onto a color gradient, and the color changes visually display the aerodynamic performance distribution on the blade surface, helping to quickly identify areas with poor performance.Based on the aerodynamic data (e.g., surface pressure, wind speed, angle of attack) of each grid node, it is mapped to a color gradient; for example, areas with higher pressure can be displayed in red, while areas with lower pressure are displayed in blue, and areas with higher wind speeds can be displayed in green, etc. For example, assuming the calculated pressure data ranges from 100 to 500 Pa, according to this range, higher pressure values ​​(e.g., 500 Pa) will be mapped to red, and lower pressure values ​​(e.g., 100 Pa) will be mapped to blue. Through the color gradient, the aerodynamic information distribution map will show which areas on the blade have good aerodynamic performance and which areas may need further design optimization or monitoring. By updating the aerodynamic information distribution map in real time and dynamically observing the changes in aerodynamic parameters on the blade surface, the wind turbine blade can be monitored in real time. The system monitors the performance of wind turbine blades; based on the update frequency of sensor data, it periodically refreshes the aerodynamic information distribution map to monitor the dynamic changes of the blades and provide timely feedback to wind turbine maintenance personnel. For example, assuming the sensor collects data once per second and updates the distribution map every 10 seconds, this means that every 10 seconds, the system will regenerate an aerodynamic information distribution map, showing changes in blade surface pressure, wind speed, and angle of attack, helping the monitoring system to promptly detect any abnormal aerodynamic phenomena on the blades. It also simulates the aerodynamic performance of wind turbine blades under different operating conditions, collecting data under these conditions to provide a basis for performance optimization; and simulates different working environments (such as different wind speeds, climate conditions, blade angles, etc.), collecting aerodynamic data under these conditions through an aerodynamic monitoring simulation model.

[0031] In an optional embodiment, an adaptive aerodynamic parameter correction model is constructed based on the blade design aerodynamic parameters, including: Obtain the aerodynamic parameters of the blade design and the actual aerodynamic requirements during operation. Based on the conservation laws of fluid mechanics and the principle of load balance of the blade structure, analyze the balance relationship between the design aerodynamic parameters and the structural load. The aerodynamic-structure coupling relationship is determined based on the equilibrium relationship and the correlation rules in the aerodynamic-structure coupling sensitive descriptor; Using real-time aerodynamic data as input and designed aerodynamic parameters as target output, an adaptive correction model for aerodynamic parameters is constructed. The adaptive correction model for aerodynamic parameters includes a deviation calculation module and a correction coefficient generation module, which are used to output the aerodynamic parameter deviation values ​​and the corresponding correction coefficients.

[0032] It should be noted that the design aerodynamic parameters are based on theoretical fluid mechanics and structural analysis, typically including parameters such as blade aerodynamic lift, drag, and flow field distribution; while the actual operational aerodynamic requirements are the aerodynamic performance the blades need to achieve in the actual working environment. The difference between these two needs to be analyzed and corrected using models. For example, suppose a wind turbine blade is designed considering factors such as wind speed and air density, resulting in ideal aerodynamic parameters (such as lift coefficient); however, in actual operation, due to factors such as wind speed variations and blade wear, the actual performance of the blades may deviate from the design parameters; fluid mechanics conservation laws (such as mass conservation, momentum conservation, etc.) and The principle of blade structure load balance (i.e., the mechanical balance of a structure under aerodynamic loads) is used to analyze the relationship between aerodynamic performance and structural loads. Specifically, design aerodynamic parameters (such as lift, airflow distribution, etc.) and structural loads (such as bending moment, shear force, etc.) are mutually influential. For example, when a wind turbine blade faces the wind, aerodynamic lift acts on the blade, resulting in certain bending moments and shear forces. These loads need to be balanced with the blade's material strength and design structure. If aerodynamic parameters deviate from expectations, it may lead to excessive blade bending or structural damage. The aerodynamic-structural coupling descriptor is a sensitivity index used to describe the mutual influence between aerodynamics and structure; it encompasses both aerodynamics and structural mechanics. The interaction between these parameters; for example, when aerodynamic parameters change, structural loads also change, and vice versa; for example, in wind turbine blade design, changes in aerodynamic parameters (such as lift) directly affect the bending deformation of the blades; through coupling descriptors, the impact of different aerodynamic changes on structural loads can be quantitatively analyzed; during actual operation, aerodynamic data of the blades (such as wind speed, airflow distribution, etc.) can be monitored in real time and input into the model. The goal of the model is to adjust the design aerodynamic parameters to match the actual operating conditions as closely as possible; the purpose of the adaptive calibration model is to optimize design parameters through real-time data feedback; for example, assuming that real-time data monitoring of the wind turbine... If the blade lift coefficient deviates, the model will automatically calculate correction parameters based on this deviation and readjust the design aerodynamic parameters to bring the blade's aerodynamic performance back to the ideal state. The difference between real-time data and design aerodynamic parameters, i.e., the deviation value, is calculated. The correction coefficient generation module then generates corresponding correction coefficients based on the deviation calculation results to adjust the design aerodynamic parameters. For example, if the actual lift coefficient of the blade is 10% lower than the design value, the deviation calculation module will output this 10% deviation value, and the correction coefficient generation module will calculate the corresponding correction coefficient based on this deviation and apply it to the design parameters to ensure that the blade achieves more accurate aerodynamic performance in the next run.

[0033] In an optional embodiment, an aerodynamic parameter adaptive correction model is used to perform deviation analysis on real-time aerodynamic data, calculate aerodynamic parameter correction coefficients, and generate a dynamic detection execution plan based on the correction coefficients and a sensitive area detection configuration table, including: Real-time aerodynamic data is input into the aerodynamic parameter adaptive correction model to calculate the aerodynamic pressure deviation and airflow velocity deviation at each monitoring point. Based on the magnitude and distribution of the deviation values, and combined with the turbulence level and vibration intensity in the aerodynamic-structure coupling sensitive descriptor, the aerodynamic parameter correction coefficients are calculated. Based on the detection priority and accuracy requirements in the sensitive area detection configuration table, adjust the sampling frequency and data filtering intensity of different areas to determine the detection resource allocation scheme; By integrating the correction coefficients, sampling frequency, and resource allocation scheme, a dynamic detection execution scheme is generated.

[0034] It should be noted that real-time aerodynamic data (such as airflow velocity and aerodynamic pressure) is input into the aerodynamic parameter adaptive correction model. By comparing the actual aerodynamic data with the design parameters, the model calculates the aerodynamic pressure deviation and airflow velocity deviation. For example, assuming the design wind speed for a wind turbine blade is 10 m / s, but during actual operation, wind speed variations cause the airflow velocity to reach 9.5 m / s, the model will calculate this 0.5 m / s deviation. The model calculates and records deviations in m / s. After calculating these deviations (e.g., changes in pressure and velocity), it combines this information with data from the aerodynamic-structure coupling sensitive descriptor (e.g., turbulence level and vibration intensity) to generate correction coefficients for aerodynamic parameters. These coefficients are used to adjust the design aerodynamic parameters to better match actual operating conditions. For example, assuming the aerodynamic lift coefficient of the blade needs adjustment due to wind speed deviation, the model calculates a correction coefficient, such as 0.98, indicating that the designed lift coefficient needs to be reduced by 2% to match the actual wind speed. Based on the calculation results of the aerodynamic-structure coupling sensitive descriptor, the model can determine the degree of impact of aerodynamic parameter changes on the structure in different regions, thereby identifying which regions require more frequent detection. The model also adjusts the sampling frequency and data filtering intensity according to accuracy requirements. For example, assuming the leading edge region of the blade is more sensitive to aerodynamic changes and therefore requires higher detection accuracy, the model will increase the detection accuracy in that region. The model adjusts the sampling frequency and increases data filtering intensity to obtain more accurate measurement results. By comprehensively considering factors such as aerodynamic changes, sensitivity, and deviation values ​​in each region, the model will rationally allocate detection resources according to detection priority and accuracy requirements. This includes adjusting the sampling frequency and data processing method for each region. For example, if the trailing edge region of the blade is relatively stable to aerodynamic changes, the model may reduce the sampling frequency in that region and concentrate resources on detecting more sensitive regions, such as the leading edge or middle region. Finally, the model will integrate the correction coefficient, sampling frequency, and resource allocation scheme to generate a dynamic detection execution scheme. This scheme will be dynamically adjusted during actual operation to ensure that aerodynamic parameters can be corrected in real time under different conditions, ensuring the optimal performance of the blade. For example, when the wind speed changes, the detection scheme will automatically adjust, increasing the monitoring of airflow velocity in the leading edge region, and correcting the lift coefficient based on real-time data to ensure the stability of the blade's aerodynamic performance under different wind speeds.

[0035] In an optional embodiment, the detection resources are aerodynamically cycle-sensingly scheduled according to a dynamic detection execution scheme to establish a sampling timing sequence synchronized with the airflow pulsation cycle, and blade aerodynamic parameter detection is performed, including: The detection tasks in the dynamic detection execution plan are analyzed, and the sensor types, quantities, and working times required for each task are calculated to obtain the sensor resource requirement set. Assess the available resources of the detection equipment, generate a sensor resource distribution map, and extract the airflow pulsation period based on the distribution map; Sampling time slots are divided according to the airflow pulsation cycle, and a sampling timing scheme synchronized with the cycle is established; According to the sampling timing scheme, the control sensor collects aerodynamic data, and the collected data is uploaded in real time through the data transmission module to perform blade aerodynamic parameter detection.

[0036] It should be noted that by analyzing each task in the dynamic detection execution plan, calculating the required sensor types, quantities, and working times, a sensor resource requirement set is finally obtained.

[0037] Step-by-step instructions: First, analyze the required sensor types and quantities for each blade aerodynamic performance testing task (e.g., monitoring wind speed, pressure, angle of attack, etc.). Then, calculate the resource requirements based on the data acquisition duration of each sensor. For example, if a task requires monitoring the pressure distribution of the blade, the sensors may need to acquire data per second, which determines the required number of sensors and their operating time. For instance, assuming the pressure distribution of the blade is to be monitored, 30 pressure sensors are deployed for each area; assuming each sensor needs to acquire data for 5 seconds, this means that 30 pressure sensors need to run simultaneously for 5 seconds, generating a sensor resource requirement set. After clarifying the sensor resources required for each task, equipment evaluation is necessary. Based on available resources, the placement and number of sensors are rationally allocated to ensure sufficient aerodynamic performance monitoring in each area. A sensor resource distribution map can then be generated, indicating the location of each sensor on the blade surface or in the airflow channel. For example, assuming the wind turbine blade surface is divided into three main areas, with a different number of sensors deployed in each area: 20 sensors in area 1, 30 in area 2, and 40 in area 3. The resource distribution map will indicate the specific locations of these sensors, ensuring real-time monitoring of the aerodynamic performance in each area. By analyzing the airflow pulsation cycle, the periodic fluctuations of airflow on the blade surface can be understood. The extraction of the airflow pulsation cycle typically depends on the number of sensors. The frequency and trend of airflow fluctuations are analyzed by collecting data over a certain period of time to determine the fluctuation cycle of parameters such as wind speed and pressure, thus helping to set subsequent sampling time slots. For example, if the data obtained by the pressure sensor shows a certain regular fluctuation within 5 seconds, Fourier transform analysis can determine that the period of the airflow pulsation is 3 seconds. This information will help to reasonably set the sampling time slots. After knowing the airflow pulsation cycle, the data collection needs to be divided into multiple time slots to ensure that the sensor can collect effective data during critical periods. The division of sampling time slots should be synchronized with the periodic fluctuations of the airflow to ensure that complete aerodynamic data information can be obtained in each cycle. For example, if the airflow pulsation cycle is 3 seconds, it can be divided into 3-second intervals. The system uses multiple sampling time slots, sampling within each slot; for example, 10 samples are taken every 3 seconds (once every 300 milliseconds) to ensure complete airflow fluctuation data is captured. Once the sampling time slots and timing scheme are determined, the system needs to control the sensor's acquisition timing according to these schemes. Simultaneously, the acquired data needs to be uploaded to the central processing system in real time for further analysis. For example, assuming the system is set to sample once every 300 milliseconds, the sensor is activated to acquire data within the sampling time slot, and the sensor uploads the acquired data to the main control system in real time via a wireless module or wired connection. For example, the wind speed and pressure data acquired by the sensor will be uploaded to the processing center via a transmission module for further aerodynamic parameter analysis and monitoring.After the sensors successfully acquire and transmit data, the system analyzes the collected aerodynamic data in real time through the data analysis module. Ultimately, it generates a detailed aerodynamic parameter report for the blades, helping to analyze the blades' performance under different operating conditions for design optimization or fault diagnosis. For example, assuming the acquired pressure and wind speed data are analyzed by the system to obtain aerodynamic performance data of the blades in different regions, the system can generate an aerodynamic performance report to provide to maintenance personnel for subsequent equipment adjustments or design optimization.

[0038] Example 2, please refer to Figure 2 This invention provides a technical solution: a method for detecting aerodynamic parameters of wind turbine blades, applicable to the aforementioned aerodynamic parameter detection system for wind turbine blades, comprising: S1. Obtain the structural design information of the target wind turbine blade, construct the blade aerodynamic topology based on the structural design information, perform aerodynamic characteristic mutation analysis on the aerodynamic topology, and determine the initial sensor deployment area. S2. Collect blade operating status data, perform time-frequency domain analysis on the operating status data, extract airflow field characteristics and structural response features, and generate aerodynamic-structural coupling sensitive descriptors. S3. Based on the initial sensor deployment area and aerodynamic-structure coupling sensitive descriptor, calculate the aerodynamic sensitivity index of each region of the blade, optimize the sensor deployment position according to the aerodynamic sensitivity index, and generate a sensitive area detection configuration table including detection priority and accuracy requirements. S4. Construct a blade aerodynamic monitoring simulation model based on the sensitive area detection configuration table, collect real-time aerodynamic data under different operating conditions through the aerodynamic monitoring simulation model, and construct an aerodynamic parameter adaptive correction model based on the blade design aerodynamic parameters. S5. Use the aerodynamic parameter adaptive correction model to perform deviation analysis on real-time aerodynamic data, calculate the aerodynamic parameter correction coefficient, and generate a dynamic detection execution plan based on the correction coefficient and the sensitive area detection configuration table. S6. Based on the dynamic detection execution plan, perform aerodynamic cycle sensing and scheduling of detection resources, establish a sampling sequence synchronized with the airflow pulsation cycle, and execute blade aerodynamic parameter detection.

[0039] The embodiments of the present invention have been described in detail above with reference to the accompanying drawings. However, the present invention is not limited thereto. Various changes can be made within the scope of knowledge possessed by those skilled in the art without departing from the spirit of the present invention.

Claims

1. A system for detecting aerodynamic parameters of wind turbine blades, characterized in that, include: The detection layout unit is used to acquire the structural design information of the target wind turbine blade, construct the blade aerodynamic topology based on the structural design information, perform aerodynamic characteristic mutation analysis on the aerodynamic topology, and determine the initial sensor deployment area. Collect blade operating status data, perform time-frequency domain analysis on the operating status data, extract airflow field characteristics and structural response features, and generate aerodynamic-structural coupling sensitive descriptors; The detection configuration unit is used to calculate the aerodynamic sensitivity index of each region of the blade based on the initial sensor deployment area and the aerodynamic-structure coupling sensitive descriptor, optimize the sensor deployment position according to the aerodynamic sensitivity index, and generate a sensitive area detection configuration table. The calibration modeling unit is used to construct a blade aerodynamic monitoring simulation model based on the sensitive area detection configuration table, collect real-time aerodynamic data under different operating conditions through the aerodynamic monitoring simulation model, and construct an aerodynamic parameter adaptive calibration model based on the blade design aerodynamic parameters. The parameter correction unit is used to perform deviation analysis on real-time aerodynamic data using the aerodynamic parameter adaptive correction model, calculate aerodynamic parameter correction coefficients, and generate a dynamic detection execution plan based on the correction coefficients and the sensitive area detection configuration table. The aerodynamic detection unit is used to perform aerodynamic cycle sensing and scheduling of detection resources according to the dynamic detection execution scheme, establish a sampling sequence synchronized with the airflow pulsation cycle, and perform blade aerodynamic parameter detection.

2. The aerodynamic parameter detection system for wind turbine blades according to claim 1, characterized in that, Based on the structural design information, an aerodynamic topology of the blade is constructed. An aerodynamic characteristic mutation analysis is performed on the aerodynamic topology to determine the initial sensor deployment area, including: Obtain structural design information, including blade length, airfoil section parameters, chord length distribution, twist angle variation, and blade material properties; Based on the structural design information, a three-dimensional aerodynamic topology of the blade is constructed, and the surface mesh of the three-dimensional aerodynamic topology of the blade is divided and key sections are marked. Based on the computational fluid dynamics model, aerodynamic simulation of the topology is performed to analyze the airflow velocity, pressure distribution and aerodynamic load variation range of different cross sections and surface regions, and obtain aerodynamic parameter variation data. The locations of aerodynamic characteristic abrupt changes are identified based on aerodynamic parameter change data. These locations include areas where the airfoil curvature change exceeds a preset threshold, critical cross sections at the leading and trailing edges, extreme aerodynamic load locations, airflow separation risk areas, and pressure gradient abrupt change areas. The areas where these abrupt change locations are located are then determined as the initial sensor deployment areas.

3. The aerodynamic parameter detection system for wind turbine blades according to claim 2, characterized in that, Time-frequency domain analysis is performed on the operational status data to extract airflow field characteristics and structural response features, generating aerodynamic-structural coupling sensitive descriptors, including: Time-frequency domain decomposition was performed on the pressure pulsation signal and vibration acceleration signal in the operating status data to obtain airflow separation characteristics and flutter mode characteristics; The turbulence intensity and pressure gradient change rate corresponding to the airflow separation characteristics, as well as the vibration amplitude and frequency characteristics corresponding to the flutter mode characteristics, are extracted to obtain the structural response characteristics; The correlation between airflow field characteristics and structural response features is analyzed, an aerodynamic-structural coupling relationship set is constructed, and an aerodynamic-structural coupling sensitive descriptor is generated based on the airflow state identifier and structural response identifier mapped from the coupling relationship set.

4. The aerodynamic parameter detection system for wind turbine blades according to claim 3, characterized in that, Based on the initial sensor deployment area and the aerodynamic-structural coupling sensitivity descriptor, the aerodynamic sensitivity index of each region of the blade is calculated, including: Extract airflow state identifiers and structural response identifiers from the aero-structure coupling sensitive descriptor; The basic sensitivity coefficient is obtained by mapping the airflow state indicator to the turbulence level, and the structural influence coefficient is obtained by analyzing the vibration intensity of the structural response indicator. The basic sensitivity coefficient and structural influence coefficient are weighted and fused according to preset weights to calculate the aerodynamic sensitivity index of each region of the blade.

5. The aerodynamic parameter detection system for wind turbine blades according to claim 4, characterized in that, Based on the aerodynamic sensitivity index, the sensor deployment locations are optimized, and a sensitive area detection configuration table, including detection priority and accuracy requirements, is generated, including: The initial sensor deployment areas are prioritized based on the aerodynamic sensitivity index to obtain a region priority sequence. Based on the aerodynamic monitoring accuracy requirements of the blades, corresponding detection accuracy requirements are matched for different priority areas, and a set of regional accuracy constraints is generated. The sensor deployment density and location are adjusted according to the regional priority sequence and regional accuracy constraint set to generate a sensitive area detection configuration table, wherein the sensitive area detection configuration table includes detection priority, accuracy requirements and sensor location.

6. The aerodynamic parameter detection system for wind turbine blades according to claim 5, characterized in that, A blade aerodynamic monitoring simulation model is constructed based on the sensitive area detection configuration table. Real-time aerodynamic data under different operating conditions is collected through this simulation model, including: Pressure sensors, wind speed sensors, and angle-of-attack sensors are deployed according to the sensor locations in the sensitive area detection configuration table, and the sensor data is associated with the grid nodes of the aerodynamic topology. The interpolation density of the aerodynamic monitoring simulation model is set, and the sensor-collected data is calibrated as known sampling points based on the radial basis function interpolation algorithm. The surface arc length distance between the point to be interpolated and the known sampling points is calculated, and the interpolation weight is determined. Based on the radial basis function interpolation algorithm, the aerodynamic information of known sampling points is interpolated to the grid nodes to be interpolated, and the surface pressure, airflow velocity and airflow angle of attack data of each node are obtained. The aerodynamic data of each node is mapped to a visual color gradient to generate a distribution map of blade aerodynamic information. The distribution map is refreshed in real time according to the preset sensor data update frequency. The dynamic changes of aerodynamic parameters on the blade surface can be observed through the distribution map to achieve real-time aerodynamic monitoring. Simulate different operating conditions and collect surface pressure, airflow velocity, and airflow angle of attack data under each condition through an aerodynamic monitoring simulation model as real-time aerodynamic data.

7. The aerodynamic parameter detection system for wind turbine blades according to claim 6, characterized in that, An adaptive aerodynamic parameter correction model is constructed based on the blade design aerodynamic parameters, including: Obtain the aerodynamic parameters of the blade design and the actual aerodynamic requirements during operation. Based on the conservation laws of fluid mechanics and the principle of load balance of the blade structure, analyze the balance relationship between the design aerodynamic parameters and the structural load. Based on the aforementioned balance relationship and the correlation rules in the aerodynamic-structure coupling sensitive descriptor, the aerodynamic-structure coupling relationship is determined; Using real-time aerodynamic data as input and designed aerodynamic parameters as target output, an adaptive correction model for aerodynamic parameters is constructed. The adaptive correction model for aerodynamic parameters includes a deviation calculation module and a correction coefficient generation module, which are used to output the aerodynamic parameter deviation values ​​and the corresponding correction coefficients.

8. The aerodynamic parameter detection system for wind turbine blades according to claim 7, characterized in that, The aerodynamic parameter adaptive correction model is used to perform deviation analysis on real-time aerodynamic data, calculate aerodynamic parameter correction coefficients, and generate a dynamic detection execution plan based on the correction coefficients and the sensitive area detection configuration table, including: Real-time aerodynamic data is input into the aerodynamic parameter adaptive correction model to calculate the aerodynamic pressure deviation and airflow velocity deviation at each monitoring point. Based on the magnitude and distribution of the deviation values, and combined with the turbulence level and vibration intensity in the aerodynamic-structure coupling sensitive descriptor, the aerodynamic parameter correction coefficients are calculated. Based on the detection priority and accuracy requirements in the sensitive area detection configuration table, adjust the sampling frequency and data filtering intensity of different areas to determine the detection resource allocation scheme; By integrating the correction coefficients, sampling frequency, and resource allocation scheme, a dynamic detection execution scheme is generated.

9. The aerodynamic parameter detection system for wind turbine blades according to claim 8, characterized in that, According to the dynamic detection execution scheme, the detection resources are aerodynamically cycle-sensing and scheduled to establish a sampling timing sequence synchronized with the airflow pulsation cycle, and blade aerodynamic parameter detection is performed, including: The detection tasks in the dynamic detection execution plan are analyzed, and the sensor types, quantities, and working times required for each task are calculated to obtain the sensor resource requirement set. Assess the available resources of the detection equipment, generate a sensor resource distribution map, and extract the airflow pulsation period based on the distribution map; Sampling time slots are divided according to the airflow pulsation cycle, and a sampling timing scheme synchronized with the cycle is established; According to the sampling timing scheme, the control sensor collects aerodynamic data, and the collected data is uploaded in real time through the data transmission module to perform blade aerodynamic parameter detection.

10. A method for detecting aerodynamic parameters of a wind turbine blade, applicable to the aerodynamic parameter detection system for a wind turbine blade as described in any one of claims 1-9, characterized in that, include: Obtain the structural design information of the target wind turbine blade, construct the blade aerodynamic topology based on the structural design information, perform aerodynamic characteristic mutation analysis on the aerodynamic topology, and determine the initial sensor deployment area; Collect blade operating status data, perform time-frequency domain analysis on the operating status data, extract airflow field characteristics and structural response features, and generate aerodynamic-structural coupling sensitive descriptors; Based on the initial sensor deployment area and the aerodynamic-structure coupling sensitive descriptor, the aerodynamic sensitivity index of each region of the blade is calculated. The sensor deployment position is optimized according to the aerodynamic sensitivity index, and a sensitive area detection configuration table including detection priority and accuracy requirements is generated. A blade aerodynamic monitoring simulation model is constructed based on the sensitive area detection configuration table. Real-time aerodynamic data under different operating conditions are collected through the aerodynamic monitoring simulation model. An aerodynamic parameter adaptive correction model is constructed based on the blade design aerodynamic parameters. The aerodynamic parameter adaptive correction model is used to perform deviation analysis on real-time aerodynamic data, calculate aerodynamic parameter correction coefficients, and generate a dynamic detection execution plan based on the correction coefficients and the sensitive area detection configuration table. According to the dynamic detection execution scheme, the detection resources are aerodynamically cycle-sensing and scheduling is performed to establish a sampling sequence synchronized with the airflow pulsation cycle, and the blade aerodynamic parameters are detected.

Images