Real-time correction control method for inclination of tower drum of wind generating set
The real-time correction method for wind turbine tower tilt, which utilizes multi-source monitoring and intelligent decision-making, solves the problems of heavy reliance on manual labor, long downtime, and low correction accuracy in existing technologies. It achieves fully automated real-time correction of tower tilt, thereby improving power generation efficiency and equipment safety.
Patent Information
- Application Number
- CN202511944001.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-22
- Publication Date
- 2026-02-13
AI Technical Summary
In existing technologies, the monitoring and correction of wind turbine tower tilt relies on manual labor, resulting in long downtime, low correction accuracy, inability to detect sudden tilting faults, and loss of power generation.
Multi-source monitoring equipment is used to collect tower data. Visual acquisition devices and sensor groups are used to obtain tower surface and mechanical information. Combined with wind speed data, a pre-trained decision model is used for real-time correction, including hierarchical collaborative judgment rules and correction level generation, and dynamic adjustment of correction execution mechanism to achieve fully automated correction.
It achieves real-time and accurate correction of tower tilt, reduces manual intervention, avoids downtime losses, improves correction accuracy and efficiency, adapts to dynamic changes in different wind field environments, and reduces misjudgment rate and energy consumption.
Smart Images
Figure CN121520125A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of wind power generation equipment control technology, specifically to a real-time correction control method for the tilt of a wind turbine tower. Background Technology
[0002] The wind turbine tower is the core load-bearing structure supporting the rotor, nacelle, and transmission system. Its verticality directly determines the operational safety and power generation efficiency of the unit. As the wind power industry expands into complex scenarios such as high capacity, high altitude, and offshore, tower heights have exceeded 150 meters and diameters have reached over 6 meters. Affected by multiple factors such as uneven foundation settlement, strong wind load impact, temperature stress cycles, and material fatigue aging, tower tilting after a certain number of years of operation is a frequent failure. Tower tilting can lead to flange connection failure, tower crack propagation, and other issues, resulting in significant economic losses such as equipment maintenance and power generation loss.
[0003] Existing technologies for monitoring and correcting tower tilt have the following shortcomings: they rely heavily on single-point tilt sensors or periodic manual inspections (such as total station measurements). Tilting sensors are susceptible to interference from tower vibrations, leading to data drift. Manual inspections require shutdown operations, with each shutdown lasting 4-8 hours, resulting in power generation loss. Furthermore, the inspection cycle is typically 3-6 months, making it impossible to detect sudden, short-term tilting faults. When excessive tilt is detected, the unit must be shut down and correction completed manually by using hydraulic jacks to push the foundation ring or tighten the flange bolts. This correction process takes 3-5 days, during which the unit is shut down, resulting in power generation loss.
[0004] Therefore, there is an urgent need for a tower tilt real-time correction control method that integrates multi-source monitoring, intelligent decision-making, real-time correction and dynamic feedback to solve the problems of high reliance on manual labor, long downtime and low correction accuracy in existing technologies. Summary of the Invention
[0005] To address the shortcomings of existing technologies, this invention aims to provide a real-time correction and control method for the tilt of wind turbine towers, solving the problems of heavy reliance on manual labor, long downtime, and low correction accuracy in existing technologies, and enabling real-time and accurate correction of tower tilt.
[0006] The basic solution provided by this invention is a method for real-time correction and control of the tilt of a wind turbine tower, comprising the following steps: S1. Collect tower operation data through multi-source monitoring equipment and read current wind speed data from the server; the multi-source monitoring equipment includes a visual acquisition device for collecting visual information of the tower surface and a sensor group for collecting mechanical and positional information of the tower. S2. Preprocess the tower operation data and wind speed data to obtain decision input features, which include wind speed feature, strain extreme value feature, position offset feature and relative displacement feature of feature points on the tower surface. S3. Input the decision input features into the pre-trained decision model. The decision model outputs a correction level of 0 to 2 based on preset hierarchical collaborative judgment rules, where level 0 indicates no correction is needed, level 1 indicates normal correction, and level 2 indicates emergency correction. The hierarchical collaborative judgment rules include: The system presets a first wind speed threshold, a second wind speed threshold, a position offset feature threshold, a strain extreme value feature threshold, a feature point relative displacement feature threshold, a first offset threshold, a second offset threshold, and a wind speed duration threshold. The first wind speed threshold is less than the second wind speed threshold, and the first offset threshold is less than the second offset threshold. By comparing the relationship between the wind speed characteristic quantity, the first wind speed threshold, and the second wind speed threshold, the operating conditions are divided into low wind speed conditions, medium wind speed conditions, and high wind speed conditions. The wind speed characteristic quantity corresponding to the low wind speed condition is less than or equal to the first wind speed threshold. The wind speed characteristic quantity corresponding to the medium wind speed condition is greater than the first wind speed threshold and less than or equal to the second wind speed threshold. The wind speed characteristic quantity corresponding to the high wind speed condition is greater than the second wind speed threshold. Calculate the deviation of each feature quantity, and then sum the deviations of the three feature quantities by weight to obtain the comprehensive deviation. The formulas for calculating the deviation of each feature quantity and the comprehensive deviation are as follows:
[0007]
[0008]
[0009]
[0010] in, The deviation of the feature point from the relative displacement feature quantity. The characteristic quantity is the relative displacement of the characteristic point. The feature threshold is the relative displacement of the feature points. The deviation of the extreme characteristic quantity of strain. For strain extreme characteristic quantity, The characteristic threshold for strain extrema; The deviation of the positional offset feature quantity. This is a position offset feature. The threshold for position offset features; For the overall deviation, , and These are the weights of the relative displacement characteristic, strain extreme value characteristic, and position offset characteristic of the feature point, respectively. When under the aforementioned low wind speed conditions, the following conditions are met: When under the aforementioned medium wind speed conditions, the following conditions are met: When under the aforementioned high wind speed conditions, the following conditions are met: When the duration of high wind speed exceeds the wind speed duration threshold, the overall deviation is amplified by a risk amplification factor, which ranges from 1.1 to 1.5. The correction level is output based on the overall deviation: when the overall deviation is less than or equal to the first offset threshold, level 0 is output; when the overall deviation is greater than the first offset threshold and less than or equal to the second offset threshold, level 1 is output; when the overall deviation is greater than the second offset threshold, level 2 is output. S4. Generate execution parameters according to the correction level, and control multiple sets of correction execution mechanisms to work in conjunction. The correction execution mechanisms are used to adjust the tilt state of the tower from three aspects: tower foundation, connection parts and center of gravity balance. S5. After calibration, the calibration effect is verified and the calibration data is stored through the multi-source monitoring device. The calibration effect includes calibration qualified and calibration unqualified. The calibration data includes calibration level, execution parameters and calibration effect. The decision model is retrained based on all calibration data in the preset period.
[0011] The principle of this invention lies in achieving fully automated real-time correction of tower tilt through a comprehensive process of multi-source data acquisition, data preprocessing, intelligent decision-making, linkage correction, and dynamic feedback. Specifically, this includes: capturing macroscopic displacement information of the tower surface using a visual acquisition device; collecting mechanical (strain) and positional microscopic data using a sensor array; and reading wind speed data from a server. The raw data undergoes preprocessing operations such as cleaning and standardization to eliminate outliers, missing values, and dimensional differences, transforming the scattered raw data into unified format decision input features (such as relative displacement features of feature points, strain extreme value features, position offset features, and wind speed features) to provide input for model decision-making. A pre-trained decision model analyzes and calculates the input features based on preset hierarchical collaborative judgment rules, outputting a correction level of 0-2. Specifically, this includes using dual wind speed thresholds (first and second wind speed thresholds) to divide the working conditions into low, medium, and high wind speed ranges, and configuring differential correction based on the differences in tilt induction factors under different working conditions. Differentiated weights are used to prioritize the inherent tilt and positional displacement characteristics of the structure under low wind speed conditions, to balance positional displacement and strain extrema under medium wind speed conditions, and to prioritize stress safety weights under high wind speed conditions. The degree of exceeding the standard is quantified by calculating the deviation between the relative displacement characteristics, strain extrema characteristics, and positional displacement characteristics of feature points and their corresponding thresholds, and a comprehensive deviation is obtained by weighting each deviation, achieving precise quantification of the tilt degree. Simultaneously, a risk amplification mechanism is added for scenarios with persistently high wind speeds to avoid safety hazards caused by the accumulation of plastic deformation. Execution parameters are generated based on the correction level, driving multiple sets of correction execution mechanisms to work collaboratively from three aspects: foundation support, pre-tightening force at connection points, and center of gravity balance, specifically offsetting tilt caused by different factors. After correction, the effect is verified in real time through multi-source monitoring equipment, and complete correction data (correction level, execution parameters, and correction effect) is stored. The decision model is retrained every preset period using historical correction data within that period, enabling the model to continuously adapt to changes in the tower's operating state.
[0012] The beneficial effects of this invention are as follows: 1. From data collection, decision-making, calibration to model optimization, no human intervention is required throughout the entire process, completely eliminating the dependence on manual inspection and calibration of existing technologies, avoiding human error, and the calibration process can be completed during normal unit operation without shutdown, which can effectively reduce power generation loss; 2. By dividing the wind speed into dual wind speed thresholds, it can accurately adapt to different wind field conditions. The differentiated weight configuration is in line with the objective law of tilt under each condition. The comprehensive deviation degree is used to achieve a quantitative assessment of the tilt degree, which can effectively reduce the misjudgment rate compared with the traditional single rule judgment. Through the risk amplification mechanism of high wind speed continuous scenarios, it effectively avoids the potential risk of elastic deformation transforming into plastic deformation under the long-term action of high wind speed. 3. Strong dynamic optimization and adaptability: Based on preset standards, the correction effect is verified and the model is retrained regularly. It can adapt to the dynamic changes of different tower types, different wind field environments and tower operating status, and the long-term operating accuracy continues to improve. 4. By using tiered threshold judgment and outputting 0~2 level correction levels, different tilt severity can be addressed to avoid over-correction or under-correction, reduce correction energy consumption and equipment wear, and extend the service life of the actuator.
[0013] Furthermore, the correction actuator includes a hydraulic module for adjusting the foundation support state, a prestressing module for adjusting the pre-tightening force of the connection parts, and a counterweight module for adjusting the tower's center of gravity. The actuator is not activated at correction level 0, the hydraulic module and prestressing module are activated at level 1, and the hydraulic module, prestressing module, and counterweight module are activated simultaneously at level 2. The hydraulic module addresses tilting caused by foundation settlement, the prestressing module addresses tilting caused by loose connections, and the counterweight module adjusts tilting caused by center of gravity shift; these three modules cover the main causes of tilting. The actuator is activated in stages according to the correction level: level 0 does not activate to save energy, level 1 activates the core module for slight tilting, and level 2 activates full-dimensional correction for severe tilting, avoiding resource waste while ensuring the correction effect covers different tilting scenarios.
[0014] Furthermore, the method includes the following step: S6, dynamically adjusting the acquisition parameters of the multi-source monitoring equipment based on wind speed data. When operating under high wind speed conditions, the exposure time of the visual acquisition device is shortened and the sampling frequency of the sensor group is increased. By adding an environmental adaptive adjustment mechanism, the problem of decreased data acquisition accuracy under harsh conditions such as high wind speeds is solved. At high wind speeds, tower vibration intensifies, the visual acquisition device is prone to image blurring, and sensor data is prone to fluctuation. By shortening the exposure time, image blurring is avoided, and the sampling frequency is increased to capture instantaneous data changes, ensuring that accurate decision input features can still be acquired even when the wind speed is too high, thereby improving stability and reliability in complex wind field environments. Attached Figure Description
[0015] Figure 1 This is a flowchart of an embodiment of a real-time tilt correction control method for wind turbine towers according to the present invention. Detailed Implementation
[0016] The following detailed description illustrates the specific implementation method: The basic implementation examples are as follows: Figure 1 As shown: A method for real-time correction and control of the tilt of a wind turbine tower, comprising the following steps: S1. Collect tower operation data through multi-source monitoring equipment and read current wind speed data from the server; the multi-source monitoring equipment includes a visual acquisition device for collecting visual information of the tower surface and a sensor group for collecting mechanical and positional information of the tower. In this embodiment, the visual acquisition device uses three sets of cameras, which are respectively installed at the top of the tower (about 5m from the top), the middle (about 45m high), and the bottom (about 10m from the foundation ring). The lenses face the outer wall of the tower and are used to capture visual information from preset reflective marking points (arranged at 2m intervals along the circumference, with a diameter of 50mm) on the tower surface. The sensor group includes strain sensors and GNSS positioning sensors. The strain sensors are installed at the flange connection (one set is arranged every 30m in height). The GNSS positioning sensors are installed at the top of the foundation ring. Real-time wind speed data is read from the wind farm operation and maintenance server via industrial Ethernet (TCP / IP protocol), with a measurement range of 0~50m / s.
[0017] S2. Preprocess the tower operation data and wind speed data to obtain decision input features, which include wind speed feature, strain extreme value feature, position offset feature and relative displacement feature of feature points on the tower surface. In this embodiment, preprocessing includes: data cleaning, using 3 Outliers were removed using criteria, and missing data were supplemented using linear interpolation. Data standardization was performed by converting the pixel coordinates of the visual image into actual physical coordinates using camera intrinsic parameters, and calculating the relative displacement of the reflective markers on the tower surface, i.e., the relative displacement of the feature points. The maximum strain value within a 10-second sliding window was extracted from the strain sensor data as the strain extremum. The difference between the actual position and the initial installation position was calculated from the GNSS positioning data as the position offset. The Min~Max normalization formula was used to map the above feature point relative displacement, strain extremum, position offset, and wind speed data to the [0,1] interval to obtain the wind speed feature quantity, strain extremum feature quantity, position offset feature quantity, and feature point relative displacement feature quantity.
[0018] S3. Input the decision input features into the pre-trained decision model. The decision model outputs a correction level of 0 to 2 based on preset hierarchical collaborative judgment rules, where level 0 indicates no correction is needed, level 1 indicates normal correction, and level 2 indicates emergency correction. The hierarchical collaborative judgment rules include: The system presets a first wind speed threshold, a second wind speed threshold, a position offset feature threshold, a strain extreme value feature threshold, a feature point relative displacement feature threshold, a first offset threshold, a second offset threshold, and a wind speed duration threshold. The first wind speed threshold is less than the second wind speed threshold, and the first offset threshold is less than the second offset threshold. By comparing the relationship between the wind speed characteristic quantity, the first wind speed threshold, and the second wind speed threshold, the operating conditions are divided into low wind speed conditions, medium wind speed conditions, and high wind speed conditions. The wind speed characteristic quantity corresponding to the low wind speed condition is less than or equal to the first wind speed threshold. The wind speed characteristic quantity corresponding to the medium wind speed condition is greater than the first wind speed threshold and less than or equal to the second wind speed threshold. The wind speed characteristic quantity corresponding to the high wind speed condition is greater than the second wind speed threshold. Calculate the deviation of each feature quantity, and then sum the deviations of the three feature quantities by weight to obtain the comprehensive deviation. The formulas for calculating the deviation of each feature quantity and the comprehensive deviation are as follows:
[0019]
[0020]
[0021]
[0022] in, The deviation of the feature point from the relative displacement feature quantity. The characteristic quantity is the relative displacement of the characteristic point. The feature threshold is the relative displacement of the feature points. The deviation of the extreme characteristic quantity of strain. For strain extreme characteristic quantity, The characteristic threshold for strain extrema; The deviation of the positional offset feature quantity. This is a position offset feature. The threshold for position offset features; For the overall deviation, , and These are the weights of the relative displacement characteristic, strain extreme value characteristic, and position offset characteristic of the feature point, respectively. When under the aforementioned low wind speed conditions, the following conditions are met: When under the aforementioned medium wind speed conditions, the following conditions are met: When under the aforementioned high wind speed conditions, the following conditions are met: When the duration of high wind speed exceeds the wind speed duration threshold, the overall deviation is amplified by a risk amplification factor, which ranges from 1.1 to 1.5. The correction level is output based on the overall deviation: when the overall deviation is less than or equal to the first offset threshold, level 0 is output; when the overall deviation is greater than the first offset threshold and less than or equal to the second offset threshold, level 1 is output; when the overall deviation is greater than the second offset threshold, level 2 is output. In this embodiment, the decision model is constructed using the C4.5 algorithm with 200 historical tilt failure data and 500 normal operation data as training sets. The preset thresholds are as follows: first wind speed threshold 0.4, second wind speed threshold 0.8, wind speed duration threshold 300s; position offset feature threshold 0.36, strain extreme value feature threshold 0.84, feature point relative displacement feature threshold 0.48; first offset threshold 0.3, second offset threshold 0.6; risk amplification factor is set to 1.3. Weighting of each operating condition: Low wind speed conditions: ; Medium wind speed conditions: ; High wind speed conditions: .
[0023] S4. Generate execution parameters according to the correction level, and control multiple sets of correction execution mechanisms to work in conjunction. The correction execution mechanisms are used to adjust the tilt state of the tower from three aspects: tower foundation, connection parts and center of gravity balance. The correction actuator includes a hydraulic module for adjusting the foundation support state, a prestressing module for adjusting the pre-tightening force of the connection parts, and a counterweight module for adjusting the tower's center of gravity. The actuator is not activated when the correction level is 0, the hydraulic module and prestressing module are activated when the correction level is 1, and the hydraulic module, prestressing module and counterweight module are activated simultaneously when the correction level is 2. In this embodiment, the hydraulic module uses an electro-hydraulic actuator, consisting of 6 sets, installed in a ring at 60° intervals on the outside of the foundation ring, with a maximum extension of 5mm. It has a built-in displacement sensor and supports automatic closed-loop control. The prestressing module uses an electric prestressing tensioning device, which corresponds one-to-one with the flange connection bolts, and the prestressing force adjustment range is 0~3000N*m. The counterweight module is a movable counterweight block (mass 500kg) installed on a circular track in the middle of the tower body, with a moving speed of 5mm / s. Execution parameter generation: Level 1 correction: The hydraulic module generates the extension amount based on the position offset characteristic (the extension amount is 1.2 times the position offset characteristic); the prestressing module increases the preload of flange bolts with strain exceeding the threshold by 10%; Level 2 correction: Based on the Level 1 execution parameters, the counterweight moves in the opposite direction of tilting, and the moving distance is 2.5 times the relative displacement characteristic of the characteristic point.
[0024] S5. After calibration, the calibration effect is verified and the calibration data is stored through the multi-source monitoring device. The calibration effect includes calibration qualified and calibration unqualified. The calibration data includes calibration level, execution parameters and calibration effect. The decision model is retrained based on all calibration data in the preset period. In this embodiment, the preset verification standards are: position offset feature value less than or equal to 0.1, strain extreme value feature value less than or equal to 0.6, and feature point relative displacement feature value less than or equal to 0.1. After correction, data is re-acquired, and the correction is judged as qualified or unqualified based on the above preset standards. If all standards are met, the correction is qualified; if any standard is not met, the correction is unqualified. The correction data is stored in an industrial database and retained for 3 years. The stored content includes the correction level, execution parameters (actuator extension and retraction, bolt preload, counterweight position), and correction results (qualified or unqualified). The preset cycle is 1 month. At the end of each month, all correction data for that month are extracted, and the decision model is retrained using the C4.5 algorithm. The threshold values and weight coefficients of each feature are updated to ensure that the model adapts to changes in the tower's operating state.
[0025] S6. Dynamically adjust the acquisition parameters of the multi-source monitoring equipment according to the wind speed data. When the high wind speed condition is in effect, shorten the exposure time of the visual acquisition device and increase the sampling frequency of the sensor group. In this embodiment, when the wind speed feature is greater than the second wind speed threshold of 0.8, i.e., when the wind speed is under high wind conditions, the exposure time of the visual acquisition device is shortened from 10ms to 5ms to avoid image blurring caused by tower vibration; the sampling frequency of the strain sensor is increased from 50Hz to 100Hz, and the sampling frequency of the GNSS positioning sensor is increased from 1Hz to 2Hz to ensure the accuracy of the decision input features.
[0026] This embodiment realizes fully automated real-time correction of tower tilt. The correction process does not require shutdown, which improves the correction efficiency and reduces the power generation loss caused by shutdown.
[0027] The above are merely embodiments of the present invention. Commonly known structures and characteristics are not described in detail here. Those skilled in the art are aware of all common technical knowledge in the field prior to the application date or priority date, are aware of all existing technologies in that field, and have the ability to apply conventional experimental methods prior to that date. Those skilled in the art can, under the guidance of this application, improve and implement this solution in combination with their own capabilities. Some typical known structures or methods should not be obstacles for those skilled in the art to implement this application. It should be noted that those skilled in the art can make several modifications and improvements without departing from the structure of the present invention. These should also be considered within the scope of protection of the present invention, and will not affect the effectiveness of the implementation of the present invention or the practicality of the patent. The scope of protection claimed in this application should be determined by the content of its claims, and the specific embodiments described in the specification can be used to interpret the content of the claims.
Claims
1. A method for real-time correction and control of the tilt of a wind turbine tower, characterized in that, Includes the following steps: S1. Collect tower operation data through multi-source monitoring equipment and read current wind speed data from the server; the multi-source monitoring equipment includes a visual acquisition device for collecting visual information of the tower surface and a sensor group for collecting mechanical and positional information of the tower. S2. Preprocess the tower operation data and wind speed data to obtain decision input features, which include wind speed feature, strain extreme value feature, position offset feature and relative displacement feature of feature points on the tower surface. S3. Input the decision input features into the pre-trained decision model. The decision model outputs a correction level of 0 to 2 based on the preset hierarchical collaborative judgment rules, where level 0 represents no correction required, level 1 represents normal correction, and level 2 represents emergency correction. The hierarchical collaborative judgment rules include: The system presets a first wind speed threshold, a second wind speed threshold, a position offset feature threshold, a strain extreme value feature threshold, a feature point relative displacement feature threshold, a first offset threshold, a second offset threshold, and a wind speed duration threshold. The first wind speed threshold is less than the second wind speed threshold, and the first offset threshold is less than the second offset threshold. By comparing the relationship between the wind speed characteristic quantity, the first wind speed threshold, and the second wind speed threshold, the operating conditions are divided into low wind speed conditions, medium wind speed conditions, and high wind speed conditions. The wind speed characteristic quantity corresponding to the low wind speed condition is less than or equal to the first wind speed threshold. The wind speed characteristic quantity corresponding to the medium wind speed condition is greater than the first wind speed threshold and less than or equal to the second wind speed threshold. The wind speed characteristic quantity corresponding to the high wind speed condition is greater than the second wind speed threshold. Calculate the deviation of each feature quantity, and then sum the deviations of the three feature quantities by weight to obtain the comprehensive deviation. The formulas for calculating the deviation of each feature quantity and the comprehensive deviation are as follows: in, The deviation of the feature point from the relative displacement feature quantity. The characteristic quantity is the relative displacement of the feature point. The feature threshold is the relative displacement of the feature points. The deviation of the extreme characteristic quantity of strain. For strain extreme characteristic quantity, The characteristic threshold for strain extrema; The deviation of the positional offset feature quantity. This is a position offset feature. The threshold for position offset features; For the overall deviation, , and These are the weights of the relative displacement characteristic, strain extreme value characteristic, and position offset characteristic of the feature point, respectively. When under the aforementioned low wind speed conditions, the following conditions are met: When under the aforementioned medium wind speed conditions, the following conditions are met: When under the aforementioned high wind speed conditions, the following conditions are met: When the duration of high wind speed exceeds the wind speed duration threshold, the overall deviation is amplified by a risk amplification factor, which ranges from 1.1 to 1.
5. The correction level is output based on the overall deviation: when the overall deviation is less than or equal to the first offset threshold, level 0 is output; when the overall deviation is greater than the first offset threshold and less than or equal to the second offset threshold, level 1 is output; when the overall deviation is greater than the second offset threshold, level 2 is output. S4. Generate execution parameters according to the correction level, and control multiple sets of correction execution mechanisms to work in conjunction. The correction execution mechanisms are used to adjust the tilt state of the tower from three aspects: tower foundation, connection parts and center of gravity balance. S5. After calibration, the calibration effect is verified and the calibration data is stored through the multi-source monitoring device. The calibration effect includes calibration qualified and calibration unqualified. The calibration data includes calibration level, execution parameters and calibration effect. The decision model is retrained based on all calibration data in the preset period.
2. The method for real-time correction and control of wind turbine tower tilt according to claim 1, characterized in that, The correction actuator includes a hydraulic module for adjusting the foundation support state, a prestressing module for adjusting the pre-tightening force of the connection parts, and a counterweight module for adjusting the tower's center of gravity. The actuator is not activated when the correction level is 0, the hydraulic module and the prestressing module are activated when the correction level is 1, and the hydraulic module, the prestressing module and the counterweight module are activated simultaneously when the correction level is 2.
3. The method for real-time correction and control of wind turbine tower tilt according to claim 2, characterized in that, It also includes the following steps: S6, dynamically adjusting the acquisition parameters of the multi-source monitoring equipment based on the wind speed data, and shortening the exposure time of the visual acquisition device and increasing the sampling frequency of the sensor group when the wind speed is high.