Vibration control method for offshore wind turbine structure

Abstract

The invention discloses a vibration control method for an offshore wind turbine structure, which comprises the following steps of: a, implementing real-time multi-source data acquisition which is realized by deploying a series of sensors on the offshore wind turbine structure and in the surrounding environment so as to obtain environmental load parameters, structural response parameters and running state parameters of the offshore wind turbine; and b, inputting the acquired data into a composite environment and structure state decoupling and feature extraction module, preprocessing the acquired data by the module, and identifying an environment load spectrum in real time. According to the vibration control method for the offshore wind turbine structure, through multi-source information fusion, online system identification and self-adaptive multi-target control, broadband, multi-mode and self-adaptive suppression of vibration of the offshore wind turbine structure is achieved, and the limitation of a traditional fixed parameter damper in a complex dynamic marine environment is overcome; complex excitation sources such as wind, waves and ocean currents and the coupling effect of the excitation sources are effectively handled.

Description

Technical Field

[0001] This invention relates to the field of wind power generation technology, specifically to a method for controlling the structural vibration of an offshore wind turbine. Background Technology

[0002] With the continued growth in global demand for renewable energy and the increasing urgency of addressing climate change, offshore wind power technology, as an efficient and clean way to obtain energy, is experiencing unprecedented rapid development. Compared to onshore wind turbines, offshore wind turbines can make full use of the more stable and robust wind energy resources in vast ocean areas, have higher power generation efficiency and require less land, and are therefore considered an important component of the future energy structure.

[0003] The patent publication number CN111637016B discloses a floating offshore wind turbine system, which includes three fixedly connected base pontoons and offshore wind power generation components installed on the base pontoons. The offshore wind power generation components are fixedly connected to the base pontoons through the wind turbine tower. The system also includes a mooring system and a structural vibration control system. The mooring system includes several sets of tensioned mooring chains fixed on the base pontoons. The structural vibration control system includes a first connecting platform installed on the wind turbine tower and several sets of semi-tensioned mooring chains fixed on the first connecting platform.

[0004] As described above, modern large offshore wind turbines have significantly increased height and rotor diameter, resulting in more flexible structures. Their natural frequencies are often lower and more concentrated. Furthermore, the external excitation sources in the marine environment are far more complex than those on land, including not only low-frequency aerodynamic loads (such as turbulent wind-induced vibrations) but also superimposed hydrodynamic loads over a wider frequency range (such as low-frequency motion and vortex-induced vibrations induced by waves and currents), as well as internal mechanical excitations such as rotor imbalance and generator torque pulsations. These excitations are multi-source, broadband, non-stationary, and mutually coupled. In this complex and dynamically changing environment… Under the excitation spectrum, traditional passive tuned mass dampers or viscoelastic dampers, due to their fixed parameters, can usually only be optimized for one or a few specific vibration modes and frequency ranges. Once the frequency range of the external excitation changes significantly, or multimodal coupled vibration occurs, their vibration reduction effect will drop significantly, and may even induce new resonance risks due to improper tuning. The reason for this is that passive systems with fixed parameters cannot intelligently adapt to the dynamic diversity of the marine environment, and their inherent response characteristics determine that they have an essential performance bottleneck when dealing with broadband, time-varying excitations. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a method for controlling the structural vibration of offshore wind turbines, which solves the problems of poor adaptability and limited vibration reduction effect of large-scale, flexible offshore wind turbines in complex and variable marine environments.

[0006] To achieve the above objectives, the present invention provides the following technical solution: A method for controlling the structural vibration of an offshore wind turbine includes the following steps:

[0007] a. Implement real-time multi-source data acquisition. Data acquisition is achieved by deploying a series of sensors on the offshore wind turbine structure and in the surrounding environment to obtain environmental load parameters, structural response parameters and operating status parameters of the offshore wind turbine.

[0008] b. Input the collected data into the composite environment and structural state decoupling and feature extraction module. The module preprocesses the collected data and identifies the environmental load spectrum and structural modal parameters in real time, assesses fatigue damage in real time, and identifies coupled vibration modes and vibration sources.

[0009] c. Input the decoupling and feature extraction results into the adaptive multi-objective vibration control strategy generation module. The module generates cooperative control commands for various types of actuators based on the online updated structural dynamics model and environmental load prediction model, combined with a multi-objective optimization algorithm.

[0010] d. Transmit the collaborative control command to the distributed multi-type actuator collaborative control module. The module drives at least two of the existing independent pitch system, semi-active damper and active mass damper of the offshore wind turbine to work together to apply precise control force or damping force to the offshore wind turbine structure to suppress structural vibration.

[0011] e. The method also includes an energy management and system health monitoring subsystem, which optimizes the energy consumption of the entire vibration control system and monitors and diagnoses the performance of the sensors, actuators and control system itself and the health status of the wind turbine structure in real time.

[0012] Preferably, the implementation of the real-time multi-source data acquisition module in step a includes the following:

[0013] A series of high-precision sensors are deployed on the offshore wind turbine structure and in its surrounding environment. The sensor array includes:

[0014] An environmental load sensor array includes redundantly configured ultrasonic wind speed and direction sensors, buoy-type wave sensors or non-contact wave sensors based on radar ranging principles, and an underwater Doppler current profiler.

[0015] The structural response sensor array includes at least a triaxial high-precision accelerometer, a bonded resistance strain gauge, a displacement sensor based on laser ranging or GPS differential positioning principle, and a precision electronic inclinometer.

[0016] The operating status sensor array includes an encoder, a generator power sensor, and a torque sensor;

[0017] The data acquisition unit is a high-speed, multi-channel synchronous acquisition device with precise timestamp marking function. All sensor data is transmitted to the central controller via industrial-grade fiber optic Ethernet or dedicated fieldbus to ensure low latency and high bandwidth of data transmission.

[0018] Preferably, the implementation of the real-time multi-source data acquisition module further includes:

[0019] The ultrasonic wind speed and direction sensor uses three-dimensional measurement technology to collect data at a sampling frequency of no less than 10 Hz, and the sampling frequency can be increased to 20 Hz under specific working conditions.

[0020] The wave sensor acquires data at a sampling frequency of no less than 2 Hz;

[0021] The underwater Doppler current profiler acquires data at a sampling frequency of no less than 1 Hz;

[0022] At least a three-axis high-precision accelerometer employs a microelectromechanical system or a piezoelectric sensor to monitor structural acceleration at a sampling frequency of not less than 100 Hz, with its measurement range set to ±2g to ±10g;

[0023] Adhesive-bonded resistance strain gauges, configured in a half-bridge or full-bridge manner, monitor structural stress changes at a sampling frequency of not less than 50 Hz. Their range can reach ±5000 micro-strains, and their linearity is better than 0.1%.

[0024] Displacement sensors based on laser ranging or differential positioning principles of Global Positioning System;

[0025] The precision electronic inclinometer monitors the static and dynamic tilt angles of a structure at a sampling frequency of no less than 10 Hz, with a measurement accuracy of ±0.005°.

[0026] The data acquisition unit ensures that the synchronization deviation of all sensor data across different channels is less than 1 microsecond;

[0027] Industrial-grade fiber optic Ethernet or dedicated fieldbus with data transmission latency of less than 1 millisecond and throughput of 1 Gbps.

[0028] Preferably, the implementation of the composite environment and structural state decoupling and feature extraction module in step b includes the following steps:

[0029] The collected raw data is preprocessed, including digital filtering, noise reduction, resampling, and time synchronization.

[0030] Real-time identification of environmental load spectrum is performed, including fast Fourier transform or wavelet transform of real-time wind speed data, spectrum analysis of real-time wave data, and separation and quantification of joint load characteristics and contribution under wind-wave-current coupling by using multivariate spectrum analysis or higher-order statistical analysis methods.

[0031] Real-time identification of structural modal parameters is performed. Operational modal analysis technology is used to process the collected structural response data in real time, and the natural frequencies, damping ratios and mode shapes of the wind turbine structure under the current operating conditions are identified online. The identification results are used to update the structural dynamics model in real time.

[0032] Real-time fatigue damage assessment is performed. Based on the identified structural strain data, combined with the rainflow counting method and the modified Miner linear cumulative damage theory, the cumulative fatigue damage degree of key structural components of the wind turbine is calculated in real time.

[0033] Coupled vibration modes and sources are identified by using signal processing methods such as cross-spectral analysis and coherence function analysis to identify the contribution of each mode in structural vibration and its main excitation sources.

[0034] Preferably, the implementation of the composite environment and structural state decoupling and feature extraction module further includes:

[0035] The digital filtering in the preprocessing uses Kalman filtering or bandpass filtering algorithms; the denoising algorithm uses wavelet transform; the resampling process unifies sensor data with different sampling frequencies to a common sampling frequency; and the time synchronization process ensures strict alignment of data on the time axis through the PTP protocol based on GPS time synchronization.

[0036] Real-time identification of environmental load spectrum combines the Hanning window function with the fast Fourier transform of wind speed data, uses the Welch method-based spectrum analysis for wave data, and adopts the bispectral analysis method to quantify the nonlinear interaction under wind-wave-current coupling.

[0037] The real-time identification of structural modal parameters adopts a method that combines random subspace identification algorithm or natural excitation technology with frequency domain decomposition. The identification process is updated every 15 minutes, or triggered when the environmental load changes significantly.

[0038] Real-time fatigue damage assessment employs the rainflow counting method in accordance with ASTM E1049-85, combined with the SN curves of specific materials and Goodman or Soderberg criteria. The assessment results are updated every 10 minutes, or immediately when an abnormal stress level is detected.

[0039] Coupled vibration mode and source identification utilizes operational deflection analysis techniques, combined with multi-point accelerometer data, to visualize the vibration modes of a structure at specific frequencies, in order to locate and understand the vibration source.

[0040] Preferably, the implementation of the adaptive multi-objective vibration control strategy generation module in step c includes the following steps:

[0041] An adaptive model predictive controller based on predictive control theory is adopted as the core control algorithm. In each control cycle, the controller uses the online updated wind turbine structural dynamics model and environmental load prediction model provided by the composite environment and structural state decoupling and feature extraction module to continuously optimize the control sequence for a future period of time.

[0042] Construct a multi-objective optimization function that comprehensively considers and optimizes the following performance indicators: minimizing the vibration acceleration amplitude at key structural points, minimizing the cumulative fatigue damage rate at key structural sections, minimizing the power generation fluctuation of the wind turbine, minimizing the energy consumption of the distributed actuator, and the stability and robustness of the entire control system.

[0043] The weighting factors are dynamically adjusted. The target weighting factors in the optimization function are dynamically adjusted in real time according to the current operating conditions, structural health status and preset safety thresholds.

[0044] Based on the optimization results, the adaptive model predictive controller generates multiple cooperative control commands for the distributed actuators. The commands include, but are not limited to, pitch angle increments, semi-active damper control voltage or current setpoints, and active mass damper thrust. The command update frequency is matched with the actuator response speed and is between 20 Hz and 100 Hz.

[0045] Preferably, the implementation of the distributed multi-type actuator collaborative control module in step d includes the following steps:

[0046] Enhanced control is achieved by utilizing the existing independent pitch control system of the offshore wind turbine. The pitch control system consists of a hydraulically or electrically driven blade pitch mechanism, and the adaptive multi-objective vibration control strategy generation module generates fine pitch angle increment commands.

[0047] A semi-active magnetorheological damper is deployed. The magnetorheological damper is based on real-time feedback of structural response and uses algorithms such as Skyhook control, LQR control or fuzzy logic control to convert the control commands generated by the adaptive control strategy module into actual current or voltage signals, driving the magnetorheological damper to generate the target damping force.

[0048] Deploy active mass dampers. The active mass damper system consists of a movable mass block, a linear motor or hydraulic actuator, a guide rail, and a feedback sensor. The active mass damper is mainly used to suppress the low-order modal vibration of the wind turbine tower and is equipped with an energy recovery device.

[0049] The distributed multi-type actuator collaborative control module implements a collaborative control strategy for different modes, different frequency ranges, and different excitation sources.

[0050] Preferably, the implementation of the energy management and system health monitoring subsystem in step e includes the following steps:

[0051] An energy management module is constructed, which includes a power management unit responsible for providing a stable and reliable power supply to all sensors, controllers, and actuators;

[0052] A system health monitoring module is constructed, which includes: sensor fault diagnosis, actuator performance evaluation, control system performance monitoring, and structural health prediction.

[0053] This invention provides a method for controlling the structural vibration of offshore wind turbines. Compared with existing technologies, it has the following advantages:

[0054] 1. This offshore wind turbine structural vibration control method, through multi-source information fusion, online system identification, and adaptive multi-objective control, achieves wide-bandwidth, multi-modal, and adaptive suppression of offshore wind turbine structural vibration. It overcomes the limitations of traditional fixed-parameter dampers in complex and dynamic marine environments, effectively copes with complex excitation sources such as wind, waves, and ocean currents and their coupling effects, enhances the operational reliability and safety of offshore wind turbines in complex and variable marine environments, especially their resistance to extreme marine environments, and provides a solid foundation for the intelligent operation and maintenance of wind turbines, reducing operation and maintenance costs.

[0055] 2. This offshore wind turbine structure vibration control method, through the coordinated control of distributed, multi-type actuators, can dynamically optimize the control strategy according to real-time operating conditions, effectively reducing the vibration response and fatigue damage accumulation rate of the tower, blades and foundation structure, significantly extending the service life of the wind turbine structure and improving operational safety.

[0056] 3. This method for controlling the structural vibration of offshore wind turbines does not require a significant increase in structural stiffness and weight, avoiding the high material costs and engineering implementation difficulties associated with traditional stiffness enhancement schemes. It aligns with the development trend of large-scale and lightweight construction, and improves the engineering feasibility of ultra-large offshore wind turbines. Attached Figure Description

[0057] Figure 1 This is a system block diagram of the offshore wind turbine structure vibration control method of the present invention;

[0058] Figure 2 This is a flowchart illustrating the vibration control method for offshore wind turbine structures according to the present invention.

[0059] Figure 3 This is an internal processing block diagram of the composite environment and structural state decoupling and feature extraction module of the present invention. Detailed Implementation

[0060] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0061] Please see Figures 1-3 The vibration control method for offshore wind turbine structures offers three technical solutions:

[0062] The first implementation method: The implementation of the real-time multi-source data acquisition module includes the following steps:

[0063] A series of sensors are deployed on the offshore wind turbine structure and in the surrounding environment to acquire accurate and real-time environmental load data, structural response data, and wind turbine operating status data;

[0064] The deployment of environmental load sensor arrays includes:

[0065] The redundant ultrasonic wind speed and direction sensors are installed on the top of the nacelle and in front of the center of the wheel hub to acquire real-time parameters such as wind speed, wind direction, wind shear coefficient and turbulence intensity. The wind speed and direction sensors collect data at a sampling frequency of not less than 10 Hz.

[0066] Buoy-type wave sensors or non-contact wave sensors based on radar ranging principles are deployed in the sea area near the wind turbine foundation or installed on the outside of the tower along the waterline to acquire real-time parameters such as wave height, wave period, wave direction, and wave spectral density. The wave sensors collect data at a sampling frequency of not less than 2 Hz.

[0067] The underwater Doppler current profiler (ADCP) is installed near the seabed of the wind turbine foundation to acquire real-time current velocity, current direction and depth profile data. The current profiler collects data at a sampling frequency of not less than 1 Hz.

[0068] The deployment of the structural response sensor array includes:

[0069] At least three-axis high-precision accelerometers, using microelectromechanical systems (MEMS) or piezoelectric sensors, are redundantly arranged at the top of the tower nacelle, multiple sections in the middle of the tower, the bottom of the tower, and key load-bearing points of the wind turbine foundation structure, to monitor the acceleration of the structure in three orthogonal directions at a sampling frequency of not less than 100 Hz.

[0070] Adhesive-mounted resistance strain gauges, in half-bridge or full-bridge configurations, are installed at blade roots, critical sections of tower walls, connections between the foundation and tower, and other stress concentration areas. They monitor changes in bending moment, torque, and axial stress under wind and wave loads at a sampling frequency of at least 50 Hz.

[0071] Displacement sensors based on laser ranging or GPS differential positioning principles, for example, installed on the top of the cabin or in the middle of the tower, monitor the displacement of the tower top and the tilt of the overall structure at a sampling frequency of not less than 10 Hz.

[0072] A precision electronic inclinometer, installed at the top of the tower and the foundation connection, monitors the static and dynamic tilt angles of the structure at a sampling frequency of no less than 10 Hz.

[0073] The deployment of operational status sensors includes:

[0074] Encoder monitors wheel hub speed;

[0075] Rotary transformer to monitor the real-time pitch angle of each blade;

[0076] Generator power sensor to monitor real-time power output;

[0077] Torque sensor monitors the torque on the generator shaft;

[0078] The data acquisition unit is a high-speed, multi-channel synchronous acquisition device with precise timestamp marking function. All sensor data is transmitted to the central controller via industrial-grade fiber optic Ethernet or dedicated fieldbus (e.g., EtherCAT or Profinet), ensuring low latency and high bandwidth for data transmission.

[0079] The implementation of the composite environment and structural state decoupling and feature extraction module includes the following steps:

[0080] The acquired raw data is preprocessed, including digital filtering (using algorithms such as Kalman filtering and bandpass filtering), noise reduction, resampling, and time synchronization processing, in order to eliminate measurement errors and high-frequency noise and ensure the synchronization of data from different sensors.

[0081] Real-time identification of environmental load spectra, including:

[0082] Perform Fast Fourier Transform (FFT) or Wavelet Transform on real-time wind speed data to identify wind energy spectrum distribution and obtain parameters such as dominant wind excitation frequency, energy density, and turbulence intensity.

[0083] Perform spectral analysis on real-time wave data to identify the dominant frequency, directionality, and energy distribution of the waves;

[0084] Multivariate spectral analysis or higher-order statistical analysis methods are used to separate and quantify the joint load characteristics and their contribution under wind-wave-current coupling.

[0085] Real-time identification of structural modal parameters is performed using Operational Modal Analysis (OMA) techniques, such as a combination of Stochastic Subspace Identification (SSI) algorithm or Natural Excitation Technique (NExT) and Frequency Domain Decomposition (FDD). The collected structural response data (acceleration, strain) are processed in real time. This process identifies the natural frequencies, damping ratios, and mode shapes of the wind turbine structure under the current operating conditions online. The identification results are used to update the structural dynamics model in real time to adapt to changes in the marine environment and the operating status of the wind turbine.

[0086] Real-time fatigue damage assessment is performed. Based on the identified structural strain data, combined with the rainflow counting method and the modified Miner linear cumulative damage theory, the cumulative fatigue damage degree of key structural components of the wind turbine is calculated in real time.

[0087] Coupled vibration modes and sources are identified. Through signal processing methods such as cross-spectral analysis and coherence function analysis, the contribution of each mode in structural vibration and its main excitation sources, such as wind load, wave load, rotor unbalance force, generator torque pulsation, etc., are identified.

[0088] By integrating multi-source information, identifying online systems, and implementing adaptive multi-objective control, broadband, multi-modal, and adaptive suppression of vibrations in offshore wind turbine structures has been achieved. This overcomes the limitations of traditional fixed-parameter dampers in complex and dynamic marine environments, effectively addressing complex excitation sources such as wind, waves, and ocean currents and their coupling effects. Through coordinated control of distributed, multi-type actuators, the control strategy can be dynamically optimized based on real-time operating conditions, effectively reducing the vibration response and fatigue damage accumulation rate of the tower, blades, and foundation structures. This significantly extends the service life of the wind turbine structure and improves operational safety.

[0089] The second implementation method differs from the first implementation method in that the implementation of the adaptive multi-objective vibration control strategy generation module includes the following steps:

[0090] An adaptive model predictive controller (AMPC) based on predictive control theory is adopted as the core control algorithm. This controller uses the wind turbine structural dynamics model and environmental load prediction model, which are updated online by the decoupling and feature extraction module of the complex environment and structural state, to continuously optimize the future control sequence in each control cycle.

[0091] Construct a multi-objective optimization function that comprehensively considers the following performance metrics:

[0092] Minimize the vibration acceleration amplitude at key structural points (e.g., tower top, blade root, foundation);

[0093] Minimize the cumulative fatigue damage rate of critical structural sections (e.g., tower bottom, blade root);

[0094] Minimize fluctuations in wind turbine power generation;

[0095] Minimize the energy consumption of distributed actuators;

[0096] The stability and robustness of the entire control system;

[0097] Implement an adaptive mechanism, which includes:

[0098] Online model parameter update: The controller dynamically updates its internal prediction model by using the modal parameters (including natural frequency, damping ratio, and mode shape) identified in real time by the decoupling and feature extraction module of the composite environment and structural state, so as to ensure that the control strategy is highly matched with the current actual dynamic characteristics of the structure.

[0099] Environmental load prediction: Combining meteorological and oceanographic forecast data with real-time sensor data provided by the real-time multi-source data acquisition module, time series prediction algorithms (e.g., Kalman filtering, autoregressive moving average model ARIMA, long short-term memory neural network LSTM model) are used to predict environmental loads such as wind speed, wave height, and ocean currents in the future short time window. The prediction results enable the controller to perform feedforward compensation and realize active early warning control.

[0100] Dynamic adjustment of weighting factors: The objective weighting factors in the optimization function are dynamically adjusted in real time according to the current operating conditions, structural health status (e.g., degree of fatigue damage) and preset safety thresholds. For example, under extreme sea conditions, the control system will prioritize structural safety and reduce the weight of vibration amplitude; under normal operation, it will take into account both power generation efficiency and energy consumption.

[0101] Based on the optimization results, AMPC generates multiple cooperative control commands for the distributed actuators. The commands include, but are not limited to, pitch angle increments, semi-active damper control voltage or current setpoints, and active mass damper thrust. The command update frequency is matched with the actuator response speed.

[0102] The implementation of the distributed multi-type actuator collaborative control module includes the following steps:

[0103] Enhanced control is achieved by utilizing the existing independent pitch control system of the offshore wind turbine. The pitch control system consists of a hydraulically or electrically driven blade pitch mechanism, which features high precision and high response speed. The fine pitch angle increment command generated by the adaptive multi-objective vibration control strategy generation module modifies the reference input of the existing pitch control system to achieve rapid and local adjustment of the blade aerodynamic force, thereby suppressing low-order bending and torsional vibrations of the tower and blades. For example, by applying differentiated pitch angle adjustments to the windward and leeward blades, the asymmetric aerodynamic load is actively changed to counteract vibration excitation.

[0104] A semi-active damper is deployed. The semi-active damper is a damper based on magnetorheological (MR) fluid and installed inside the tower (e.g., at the top of the tower, the middle of the tower, or at the connection of the underwater foundation). It can be used as a bypass or series damper. The MR damper dynamically changes the damping force by precisely controlling the coil current and adjusting the yield strength of the magnetorheological fluid in real time. The control command is the damping force or current set value. This damper can provide adjustable damping over a wide frequency range and is suitable for suppressing the bending and torsional vibrations of the tower. The MR damper is based on real-time feedback of structural response (e.g., displacement, velocity) and uses algorithms such as Skyhook control, LQR control, or fuzzy logic control to convert the control command generated by the adaptive control strategy module into actual current or voltage signals to drive the MR damper to generate the target damping force.

[0105] Deploying an active mass damper (AMD) system, which consists of a movable mass, a linear motor or hydraulic actuator, a guide rail, and feedback sensors, is installed inside the nacelle or on top of the tower. The linear motor or hydraulic actuator generates thrust commands based on the adaptive multi-objective vibration control strategy module, driving the mass to generate an inertial force opposite to the structural vibration, thereby counteracting the structural vibration. AMD is mainly used to suppress low-order (e.g., first-order and second-order) modal vibrations of the wind turbine tower. Its effectiveness is related to the stroke and mass of the mass. AMD is preferably equipped with an energy recovery device, such as converting the kinetic energy generated by the movement of the mass into electrical energy for storage or reuse.

[0106] The distributed multi-type actuator collaborative control module implements a collaborative control strategy, which optimizes the allocation and collaborative operation of various actuators for different modes, frequency ranges, and excitation sources. For example, the pitch system is mainly responsible for the active suppression and power regulation of low-frequency large-amplitude vibrations, the MR damper is responsible for the fine damping and local energy dissipation of mid-to-high frequency vibrations, and the AMD is responsible for the large-scale vibration suppression of specific low-order modes. The collaborative strategy achieves the overall optimal vibration reduction effect by coordinating the response time, range of action, and control capability of each actuator, and avoids conflicts or redundancy between different actuators.

[0107] The third implementation method differs from the second implementation method in that the implementation of the energy management and system health monitoring subsystem includes the following steps:

[0108] An energy management module is constructed, which includes a power management unit responsible for providing stable and reliable power to all sensors, controllers and actuators. The adaptive multi-objective vibration control strategy generation module takes actuator energy consumption as one of the optimization objectives during the optimization process to minimize overall energy consumption. The energy management module integrates a high-capacity supercapacitor or battery pack as a short-term backup power source and sets up an energy recovery unit in the AMD system to convert some of the vibration energy into electrical energy for storage, thereby improving the system's energy efficiency and sustainability.

[0109] The system health monitoring module includes:

[0110] Sensor fault diagnosis: By using methods such as sensor redundancy configuration, data cross-validation, and statistical pattern recognition (e.g., principal component analysis PCA, support vector machine SVM), the operating status of each sensor is monitored in real time, faulty sensors are identified and isolated, and the system is switched to a backup sensor or a data reconstruction algorithm is used to make up for missing data.

[0111] Actuator performance evaluation: Real-time monitoring of actuators (including pitch system, MR damper, AMD) control inputs (e.g., current, voltage, command force), output responses (e.g., blade angle, damping force, mass displacement), and internal operating parameters (e.g., hydraulic pressure, motor temperature) to evaluate their working efficiency, response speed, and whether there is mechanical wear or failure, and to provide timely warnings;

[0112] Control system performance monitoring: Real-time evaluation of the vibration reduction effect of the vibration control system (e.g., vibration amplitude reduction rate, fatigue damage reduction rate) and comparison with preset targets. If the performance deteriorates, an adaptive adjustment or fault diagnosis process is triggered.

[0113] Structural health prediction: Based on fatigue damage assessment results and historical data provided by the composite environment and structural state decoupling and feature extraction module, combined with machine learning or deep learning models, the remaining life of key parts of the wind turbine structure is predicted, providing a scientific basis for maintenance and repair.

[0114] While ensuring excellent vibration reduction performance, the system effectively reduces energy consumption and improves economic efficiency and sustainable operation by optimizing control algorithms and energy management.

[0115] It does not require a significant increase in structural stiffness and weight, avoiding the high material costs and engineering implementation difficulties brought about by traditional stiffness enhancement schemes. It conforms to the development trend of large-scale and lightweight, improves the engineering feasibility of ultra-large offshore wind turbines, enhances the operational reliability and safety of offshore wind turbines in complex and variable marine environments, especially their resistance to extreme marine environments, and provides a solid foundation for the intelligent operation and maintenance of wind turbines, reducing operation and maintenance costs.

[0116] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0117] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for controlling structural vibration of offshore wind turbines, characterized in that: Includes the following steps: a. Implement real-time multi-source data acquisition. Data acquisition is achieved by deploying a series of sensors on the offshore wind turbine structure and in the surrounding environment to obtain environmental load parameters, structural response parameters and operating status parameters of the offshore wind turbine. b. Input the collected data into the composite environment and structural state decoupling and feature extraction module. The module preprocesses the collected data and identifies the environmental load spectrum and structural modal parameters in real time, assesses fatigue damage in real time, and identifies coupled vibration modes and vibration sources. c. Input the decoupling and feature extraction results into the adaptive multi-objective vibration control strategy generation module. The module generates cooperative control commands for various types of actuators based on the online updated structural dynamics model and environmental load prediction model, combined with a multi-objective optimization algorithm. d. Transmit the collaborative control command to the distributed multi-type actuator collaborative control module. The module drives at least two of the existing independent pitch system, semi-active damper and active mass damper of the offshore wind turbine to work together to apply precise control force or damping force to the offshore wind turbine structure to suppress structural vibration. e. The method also includes an energy management and system health monitoring subsystem, which optimizes the energy consumption of the entire vibration control system and monitors and diagnoses the performance of the sensors, actuators and control system itself and the health status of the wind turbine structure in real time.

2. The method for controlling structural vibration of an offshore wind turbine according to claim 1, characterized in that: The implementation of the real-time multi-source data acquisition module in step a includes the following: A series of high-precision sensors are deployed on the offshore wind turbine structure and in its surrounding environment. The sensor array includes: An environmental load sensor array includes redundantly configured ultrasonic wind speed and direction sensors, buoy-type wave sensors or non-contact wave sensors based on radar ranging principles, and an underwater Doppler current profiler. The structural response sensor array includes at least a triaxial high-precision accelerometer, a bonded resistance strain gauge, a displacement sensor based on laser ranging or GPS differential positioning principle, and a precision electronic inclinometer. The operating status sensor array includes an encoder, a generator power sensor, and a torque sensor; The data acquisition unit is a high-speed, multi-channel synchronous acquisition device with precise timestamp marking function. All sensor data is transmitted to the central controller via industrial-grade fiber optic Ethernet or dedicated fieldbus to ensure low latency and high bandwidth of data transmission.

3. The method for controlling structural vibration of an offshore wind turbine according to claim 2, characterized in that: The implementation of the real-time multi-source data acquisition module further includes: The ultrasonic wind speed and direction sensor uses three-dimensional measurement technology to collect data at a sampling frequency of no less than 10 Hz, and the sampling frequency is increased to 20 Hz under specific working conditions. The wave sensor acquires data at a sampling frequency of no less than 2 Hz; The underwater Doppler current profiler acquires data at a sampling frequency of no less than 1 Hz; At least a three-axis high-precision accelerometer employs a microelectromechanical system or a piezoelectric sensor to monitor structural acceleration at a sampling frequency of not less than 100 Hz, with its measurement range set to ±2g to ±10g; Adhesive-bonded resistance strain gauges, configured in a half-bridge or full-bridge manner, monitor structural stress changes at a sampling frequency of not less than 50 Hz. Their range can reach ±5000 micro-strains, and their linearity is better than 0.1%. Displacement sensors based on laser ranging or differential positioning principles of Global Positioning System; A precision electronic inclinometer monitors the static and dynamic tilt angles of a structure at a sampling frequency of no less than 10 Hz. The data acquisition unit ensures that the synchronization deviation of all sensor data across different channels is less than 1 microsecond; Industrial-grade fiber optic Ethernet or dedicated fieldbus with data transmission latency of less than 1 millisecond and throughput of up to 1 Gbps.

4. The method for controlling structural vibration of an offshore wind turbine according to claim 1, characterized in that: The implementation of the composite environment and structural state decoupling and feature extraction module in step b includes the following steps: The collected raw data is preprocessed, including digital filtering, noise reduction, resampling, and time synchronization. Real-time identification of environmental load spectrum is performed, including fast Fourier transform or wavelet transform of real-time wind speed data, spectrum analysis of real-time wave data, and separation and quantification of joint load characteristics and contribution under wind-wave-current coupling by using multivariate spectrum analysis or higher-order statistical analysis methods. Real-time identification of structural modal parameters is performed. Operational modal analysis technology is used to process the collected structural response data in real time, and the natural frequencies, damping ratios and mode shapes of the wind turbine structure under the current operating conditions are identified online. The identification results are used to update the structural dynamics model in real time. Real-time fatigue damage assessment is performed. Based on the identified structural strain data, combined with the rainflow counting method and the modified Miner linear cumulative damage theory, the cumulative fatigue damage degree of key structural components of the wind turbine is calculated in real time. Coupled vibration modes and sources are identified by using cross-spectral analysis, coherence function analysis and other signal processing methods to identify the contribution of each mode in structural vibration and its main excitation sources.

5. The method for controlling structural vibration of an offshore wind turbine according to claim 4, characterized in that: The implementation of the composite environment and structural state decoupling and feature extraction module further includes: The digital filtering in the preprocessing uses Kalman filtering or bandpass filtering algorithms; the denoising algorithm uses wavelet transform; the resampling process unifies sensor data with different sampling frequencies to a common sampling frequency; and the time synchronization process ensures strict alignment of data on the time axis through the PTP protocol based on GPS time synchronization. Real-time identification of environmental load spectrum combines the Hanning window function with the fast Fourier transform of wind speed data, uses the Welch method-based spectrum analysis for wave data, and adopts the bispectral analysis method to quantify the nonlinear interaction under wind-wave-current coupling. The real-time identification of structural modal parameters adopts a method that combines random subspace identification algorithm or natural excitation technology with frequency domain decomposition. The identification process is updated every 15 minutes, or triggered when the environmental load changes significantly. Real-time fatigue damage assessment employs the rainflow counting method in accordance with ASTM E1049-85, combined with the SN curves of specific materials and Goodman or Soderberg criteria. The assessment results are updated every 10 minutes, or immediately when an abnormal stress level is detected. Coupled vibration mode and source identification utilizes operational deflection analysis techniques, combined with multi-point accelerometer data, to visualize the vibration modes of a structure at specific frequencies, in order to locate and understand the vibration source.

6. The method for controlling structural vibration of an offshore wind turbine according to claim 1, characterized in that: The implementation of the adaptive multi-objective vibration control strategy generation module in step c includes the following steps: An adaptive model predictive controller based on predictive control theory is adopted as the core control algorithm. In each control cycle, the controller uses the online updated wind turbine structural dynamics model and environmental load prediction model provided by the composite environment and structural state decoupling and feature extraction module to continuously optimize the control sequence for a future period of time. Construct a multi-objective optimization function that comprehensively considers and optimizes the following performance indicators: minimizing the vibration acceleration amplitude at key structural points, minimizing the cumulative fatigue damage rate at key structural sections, minimizing the power generation fluctuation of the wind turbine, minimizing the energy consumption of the distributed actuator, and the stability and robustness of the entire control system. The weighting factors are dynamically adjusted. The target weighting factors in the optimization function are dynamically adjusted in real time according to the current operating conditions, structural health status and preset safety thresholds. Based on the optimization results, the adaptive model predictive controller generates multiple cooperative control commands for the distributed actuators. The commands include, but are not limited to, pitch angle increments, semi-active damper control voltage or current setpoints, and active mass damper thrust. The command update frequency is matched with the actuator response speed and is between 20 Hz and 100 Hz.

7. The method for controlling structural vibration of an offshore wind turbine according to claim 1, characterized in that: The implementation of the distributed multi-type actuator cooperative control module in step d includes the following steps: Enhanced control is achieved by utilizing the existing independent pitch control system of the offshore wind turbine. The pitch control system consists of a hydraulically or electrically driven blade pitch mechanism, and the adaptive multi-objective vibration control strategy generation module generates fine pitch angle increment commands. Deploy a semi-active magnetorheological damper. The magnetorheological damper is based on real-time feedback of structural response and adopts Skyhook control, LQR control or fuzzy logic control algorithm to convert the control command generated by the adaptive control strategy module into actual current or voltage signal to drive the magnetorheological damper to generate the target damping force. Deploy active mass dampers. The active mass damper system consists of a movable mass block, a linear motor or hydraulic actuator, a guide rail, and a feedback sensor. The active mass damper is used to suppress the low-order modal vibration of the wind turbine tower and is equipped with an energy recovery device. The distributed multi-type actuator collaborative control module implements a collaborative control strategy for different modes, different frequency ranges, and different excitation sources.

8. The method for controlling structural vibration of an offshore wind turbine according to claim 1, characterized in that: The implementation of the energy management and system health monitoring subsystem in step e includes the following steps: An energy management module is constructed, which includes a power management unit responsible for providing a stable and reliable power supply to all sensors, controllers, and actuators; A system health monitoring module is constructed, which includes: sensor fault diagnosis, actuator performance evaluation, control system performance monitoring, and structural health prediction.

Images