CN-122015780-A - Wave data acquisition device, monitoring system and method for offshore floating wind turbine generator

CN122015780ACN 122015780 ACN122015780 ACN 122015780ACN-122015780-A

Abstract

The invention discloses a wave data acquisition device, a monitoring system and a method of an offshore floating wind turbine, aiming at the problems of non-contact wave sensor measurement deviation caused by floating platform movement and fixed window statistics hysteresis in a traditional wave monitoring method, a 'sensor-inertial navigation module' collaborative fusion framework is constructed, six-degree-of-freedom motion interference is compensated in real time by utilizing inertial navigation data, an adaptive time sliding window mechanism based on wave height variation coefficients is provided, statistical time length and sampling frequency are dynamically adjusted, prediction results of a plurality of prediction models are fused, and effective wave height, period, wave direction and 95% confidence interval are output through triple verification of error threshold values, extremum ranges and physical consistency. The wave monitoring system and the wave monitoring method can remarkably improve accuracy, instantaneity and robustness of wave monitoring, and provide reliable data support for safe operation, control load reduction, operation and maintenance decision and hydrologic environment analysis of the unit.

Inventors

SHANG WEI
CHEN BAOKANG
WEI YUFENG
CHEN JUNLIN
HAN LEI
HE JIANBIN
XU NIANDONG

Assignees

明阳智慧能源集团股份公司

Dates

Publication Date: 20260512
Application Date: 20251209

Claims (10)

1. The wave data acquisition device of the offshore floating wind turbine is characterized by comprising a fixed bracket, a turnover bracket, a non-contact wave sensor integrated with an inertial navigation module and a sensor fixing piece; Wherein, the fixed bracket is fixed with the guardrail on the tower foundation guardrail platform; The overturning bracket is rotationally connected with the fixed bracket; The sensor fixing piece is arranged on the overturning bracket; the non-contact wave sensor integrated with the inertial navigation module is arranged on the sensor fixing piece and distributed in a regular triangle shape, and the position of the non-contact wave sensor is driven by the overturning bracket to be adjusted.
2. The wave data acquisition device of the offshore floating wind turbine of claim 1, wherein the turnover bracket comprises a turnover main frame, a telescopic frame and a telescopic support arm assembly; Wherein, the turnover main frame is provided with a chute; the telescopic frame is arranged in the chute and stretches along the extending direction of the chute; The telescopic supporting arm assembly is connected between the turnover main frame and the telescopic frame.
3. The wave data monitoring system of the offshore floating wind turbine is characterized by comprising the wave data acquisition device, the data processing module and the monitoring module of the offshore floating wind turbine of claim 1 or 2; the data processing module processes the data acquired by the wave data acquisition device of the offshore floating wind turbine, transmits the processed data to the monitoring module, and monitors the ocean wave data by the monitoring module; the data processing module comprises a motion compensation unit, a data filtering unit, a self-adaptive statistical unit, a prediction unit, a fusion checking unit and a data output unit; the motion compensation unit converts an observed value of the non-contact wave sensor under a carrier coordinate system into a global coordinate system by adopting an Euler angle transformation method, so that coupling interference of six-degree-of-freedom motion of the floating platform on wave measurement is stripped, and real wave data under the global coordinate system is obtained; the data filtering unit is used for removing local abnormal points caused by sensor noise, sea surface stray echoes or compensation residual errors; the self-adaptive statistical unit is used for calculating a wave height variation coefficient CV and dynamically adjusting sliding window parameters; the prediction unit is used for training a prediction model based on historical wave data and outputting future wave prediction data through the trained prediction model; the fusion checking unit is used for checking the consistency of the statistical result and the prediction result under multiple constraint conditions; the data output unit is used for outputting real-time wave parameters, statistical characteristic reports and abnormal alarm information; The monitoring module comprises a remote wind power station service center and a local wind turbine generator control system.
4. The wave data monitoring method for the offshore floating wind turbine is characterized by comprising the following steps of: s1, acquiring original wave observation data containing motion interference of a floating type platform and six-degree-of-freedom pose data of the floating type platform through a wave data acquisition device of the offshore floating type wind turbine; S2, converting an observation value of the non-contact wave sensor under a carrier coordinate system into a global coordinate system by adopting an Euler angle transformation method through a motion compensation unit, so as to peel off coupling interference of six-degree-of-freedom motion of the floating platform on wave measurement and obtain real wave data under the global coordinate system; S3, removing local abnormal points caused by instantaneous interference or compensation residual errors generated by sensor noise and wave breaking in the real wave data through a data filtering unit to obtain purified wave data; S4, aiming at the purified wave data, calculating a wave height variation coefficient CV through a self-adaptive statistical unit, dynamically adjusting sliding window parameters, and extracting historical wave characteristics; s5, training a prediction model based on historical wave characteristics through a prediction unit and predicting future wave prediction data; S6, carrying out multi-constraint condition consistency check on the statistical result obtained in the step S4 and the prediction result obtained in the step S5 through a fusion check unit; S7, outputting real-time wave parameters, statistical characteristic reports and abnormal alarm information through a data output unit; and S8, monitoring ocean wave data through a monitoring module based on the real-time wave parameters, the statistical characteristic report and the abnormal alarm information.
5. The method for monitoring wave data of an offshore floating wind turbine of claim 4, wherein step S2 comprises: S2-1, constructing a rotation matrix from a carrier coordinate system to a global coordinate system : Wherein, the Representing the roll angle of the floating platform, Representing the pitch angle of the floating platform, Representing a heading angle of the floating platform; s2-2, combined rotation matrix Calculating target position in global coordinate system : Wherein, the Is the global position of the carrier and, For the position offset of the sensor in the carrier coordinate system, As raw measurement of a non-contact wave sensor, The displacement change of the non-contact wave sensor caused by the movement of the floating platform is generated; S2-3, compensating for six-degree-of-freedom motion of the floating platform through reverse operation, and obtaining real wave data under a global coordinate system after compensation: Wherein, the In order to compensate for the effective wave height after compensation, For the wave height of the original observation, Is the heave amount of the floating platform and is based on the target position Obtaining; in order to compensate for the wave period after the compensation, The wave period obtained for the original observation; In order to compensate for the wave direction after the compensation, For the direction of the wave as originally observed, Is the initial heading angle.
6. The method for monitoring wave data of an offshore floating wind turbine of claim 4, wherein step S3 is performed by filtering by the following filtering mechanism: S3-1, decomposing the compensated wave height data into a low-frequency signal corresponding to a real wave component and a high-frequency signal corresponding to a noise component by adopting a db4 wavelet basis; S3-2, by self-adapting threshold function The high-frequency coefficient is subjected to threshold processing, and effective signal components are reserved; Is the standard deviation of the noise, which is the standard deviation of the noise, Data length; S3-3, calculating the average value of the denoised data sequence And standard deviation of noise ; S3-4, based on normal distribution characteristics, will exceed Data points of the range are marked as outliers; S3-5, if the number of the continuous abnormal points is not more than 5, replacing the continuous abnormal points by adopting a linear interpolation method, if the number of the continuous abnormal points is not more than 5, judging that the data segment is subjected to strong interference or sensor faults, marking the data segment as an invalid data segment, automatically skipping the data segment during subsequent statistical analysis and feature extraction, and not participating in calculation; s3-6, outputting the purified wave data.
7. The method for monitoring wave data of an offshore floating wind turbine of claim 4, wherein step S4 comprises: Continuously calculating wave height variation coefficients of the last 3 sliding windows : Wherein, the As the standard deviation of the wave height, Is the average value of wave height; If CV is more than 0.3, judging that the waves are severe, switching the statistical duration of wave history data to 5-8 minutes, wherein the sampling frequency is 6-10Hz, capturing high-frequency dynamic characteristics, if CV is less than or equal to 0.2, switching the statistical duration of wave history data to 15-20 minutes, the sampling frequency is 3-5Hz, reflecting stable statistical characteristics, and if CV is less than or equal to 0.2, switching the statistical duration of wave history data to 20-30 minutes, and the sampling frequency is 3-5Hz; And optimizing the statistical duration and sampling frequency of wave history data by combining the sea area wave type and the unit operation scene: For sea areas where stormy waves are dominant: The method comprises the steps of monitoring a machine set operation scene, wherein the wave history data statistics time is 15-20 minutes, the sampling frequency is 3-5Hz, the wave history data statistics time is 5-8 minutes, the sampling frequency is 6-10Hz if the machine set operation scene is extreme weather early warning, the wave history data statistics time is 20-25 minutes, the sampling frequency is 5-10Hz if the machine set operation scene is operation window period evaluation, the wave history data statistics time is 60 minutes and the sampling frequency is 3-10Hz if the machine set operation scene is long-term data accumulation; For sea areas where swells predominate: The method comprises the steps of monitoring a machine set operation scene, wherein the wave history data statistics time is 18-20 minutes, the sampling frequency is 3-5Hz, the wave history data statistics time is 8-10 minutes, the sampling frequency is 6-10Hz if the machine set operation scene is an extreme weather early warning, the wave history data statistics time is 25-30 minutes, the sampling frequency is 5-10Hz if the machine set operation scene is an operation window period assessment, the wave history data statistics time is 60 minutes and the sampling frequency is 3-10Hz if the machine set operation scene is a long-term data accumulation; For sea areas where mixed waves predominate: The method comprises the steps of monitoring a machine set operation scene normally, wherein the wave history data statistics time is 16-18 minutes, the sampling frequency is 3-5Hz, the wave history data statistics time is 6-9 minutes, the sampling frequency is 6-10Hz if the machine set operation scene is extreme weather early warning, the wave history data statistics time is 22-28 minutes, the sampling frequency is 5-10Hz if the machine set operation scene is operation window period evaluation, the wave history data statistics time is 60 minutes and the sampling frequency is 3-10Hz if the machine set operation scene is long-term data accumulation; After the optimization adjustment is completed, the historical wave characteristics are extracted: Effective wave height Taking an arithmetic average value of 1/3 of the wave height of the large wave in the window: Wherein, the Are all waves within the window of the device, N is the total wave number in the window; Maximum wave height The maximum value of wave height in a window; Wave height standard deviation Reflecting the dispersion degree of wave height distribution: is the first in the window The number of waves to be transmitted by the optical system, Is the arithmetic average value of the wave height in the window; Peak period Converting wave data into a frequency domain through Fourier transformation, and extracting a period corresponding to the maximum frequency of energy: is the peak frequency of the energy spectrum; Average period Arithmetic mean of all wave periods within the window: is the first A period of the individual waves; Coefficient of periodic variation : Is the periodic standard deviation; Dominant wave direction Counting the mode of wave direction distribution in the window, namely the wave direction with highest occurrence frequency; Standard deviation of wave direction distribution Reflecting the concentration degree of the wave direction, The smaller the wave direction, the more stable.
8. The method for monitoring wave data of the offshore floating wind turbine generator according to claim 7, wherein the adaptive statistics unit satisfies the following constraint conditions when performing data statistics: minimum sample size a single set of statistics contains at least 30 complete wave cycles; the identification accuracy rate is that the peak-valley identification accuracy rate of the zero crossing method after adjustment is more than or equal to 90 percent; and (3) system synchronism, namely matching the statistical period with the updating frequency of the wind turbine generator control system data.
9. The method for monitoring wave data of the offshore floating wind turbine generator according to claim 7, wherein in the step S5, different prediction models are selected for prediction according to different time sequence prediction requirements; Wherein, the Selecting a TFT prediction model for long duration windows of 30 minutes or more and long period predictions of 1 hour or more; Aiming at nonlinear complex fluctuation prediction of a short-time long window of 5-10 minutes, nonlinear complex fluctuation, namely a wave height variation coefficient CV >0.3, selecting a TCN prediction model; aiming at stationary characteristic data, namely data with wave height variation coefficient CV less than or equal to 0.2, an N-BEATS prediction model is selected, and the stationary sequence prediction with high automation is realized by performing basis function decomposition through deep learning.
10. The method for monitoring wave data of an offshore floating wind turbine of claim 9, wherein step S6 comprises: S6-1, associating the historical wave characteristics of the latest window with predicted data to construct a 'history-future' continuous characteristic sequence, wherein the dimension is 1X (M X2), and M is a single window characteristic number; S6-2, calculating the deviation between the predicted value and the average value of the historical wave characteristics of the near 3 windows, and performing error threshold verification, if the deviation of the wave height is less than or equal to 20%, the deviation of the period is less than or equal to 15% and the deviation of the wave direction is less than or equal to 10, entering a step S6-3, otherwise, marking that the error exceeds the standard, and returning to the step S4; s6-3, calculating standard deviation of the multi-model prediction result to carry out consistency check, if the standard deviation is less than or equal to 5%, judging that the prediction is stable, otherwise, marking that the prediction is unstable, and returning to the step S4.

Description

Wave data acquisition device, monitoring system and method for offshore floating wind turbine generator Technical Field The invention relates to the technical field of data monitoring, in particular to a wave data acquisition device, a wave data monitoring system and a wave data monitoring method for an offshore floating wind turbine. Background With the rapid development of deep sea wind power development, an offshore floating wind turbine has become an important technical route for developing deep sea wind energy resources (the offshore floating wind turbine comprises a floating platform and a wind turbine mounted on the floating platform, the wind turbine comprises blades, a hub, a pitch system, a cabin and a tower, and a tower foundation guardrail platform is arranged on the tower), but the floating platform is easy to generate great motion in a complex marine environment, and risks such as loose mooring ropes, structural fatigue and even system instability can be caused. In addition, the operation, the installation and the overhaul operation of the offshore floating wind turbine are highly dependent on real-time sea condition conditions, in particular to wave parameters. If the accurate and real-time wave monitoring means are lacking, the operation and maintenance efficiency can be obviously affected, and the operation safety can be possibly endangered. Traditional weather forecast is generally based on average data of a large-range sea area, wave characteristics (such as wave height, period and wave direction) of a local water area of a wind power plant are difficult to accurately reflect, so that an operation and maintenance ship is forced to return to voyage due to the fact that actual sea conditions exceed an operation window after arriving at the site, and invalid voyage and economic cost are increased. Traditional contact wave monitoring equipment (such as a wave buoy) is easily affected by seawater corrosion and biological attachment, and risks such as capsizing, cable breakage and the like exist, so that the device is not suitable for long-term stable monitoring of a floating platform. The measurement data of the contact sensor arranged on the floating platform is a mixed result of the motion of the floating platform and the relative motion caused by waves, and the real characteristics of the waves cannot be truly reflected. The non-contact wave sensor is mainly based on the radar ranging principle, and inverts wave elevation information by transmitting signals to the sea surface and receiving echoes. The technology can carry out non-invasive detection on the sea surface with a certain distance outside the floating platform, directly acquire an undisturbed original wave field in front of the platform, and provide pure wave input data. The prior art still faces the following challenges: 1. the non-contact wave sensor is not convenient enough in installation mode, high-altitude operation risk is high, safety of operation and maintenance personnel is threatened, equipment failure downtime is long, and clearance distance is difficult to flexibly adjust. 2. The offshore floating wind turbine generator has unique floating characteristics, and six-degree-of-freedom motions such as heave, longitudinal and transverse shaking and the like can be generated under the action of waves. The non-contact wave sensor is usually fixed on a platform of the offshore floating wind turbine generator and moves synchronously with the platform, so that the measurement accuracy of the non-contact wave sensor is easily interfered by the movement of the platform, and the real parameters of waves are difficult to obtain. 3. Traditional wave monitoring relies on historical data for a certain period of time to statistically output wave parameters, and there is an inherent time delay. The offshore floating wind turbine needs to adjust actions such as yaw, pitch and the like according to real-time wave states, a control command is disjointed from an actual wave working condition due to a lagged statistical value, and when extreme sudden waves are faced, the lagged statistical result cannot be early-warned in time, and the best opportunity for avoiding is possibly misplaced by go wrong. In addition, the time length selection of the historical data can also directly influence the accuracy and reliability of the monitoring result. Disclosure of Invention The invention aims to overcome the defects of the prior art and provides a wave data acquisition device, a wave data monitoring system and a wave data monitoring method for an offshore floating wind turbine. In order to achieve the above purpose, the technical scheme provided by the invention is as follows: The wave data acquisition device of the offshore floating wind turbine comprises a fixed bracket, a turnover bracket, a non-contact wave sensor integrated with an inertial navigation module and a sensor fixing piece; Wherein, the fixed bracket is fixed with the guardrail on the