Search

CN-121976921-A - Real-time measuring, collecting, analyzing and alarming processing method for swing amplitude of fan tower

CN121976921ACN 121976921 ACN121976921 ACN 121976921ACN-121976921-A

Abstract

The invention discloses a method for measuring, collecting, analyzing and alarming swing amplitude of a fan tower in real time, which relates to the technical field of wind power monitoring and aims to solve the technical problems of data distortion, inaccurate prediction alarming and lack of pertinence in interference caused by mooring and tower swing coupling in the existing typhoon environment, and comprises the following steps of S1, synchronously collecting tower swing data, mooring system tension data and typhoon environment characteristic data, identifying instantaneous loosening and tightening states of a mooring cable, constructing a mooring tension-tower swing dynamic coupling model adapting to the whole stage of typhoon, and carrying out decoupling treatment to obtain tower pure swing data; S2, constructing a staged multi-parameter mapping relation prediction model based on a typhoon full-stage load database, and predicting tower barrel swing peaks and variation trends of different stages of typhoons. The method has the advantages of accurately acquiring pure swing data, predicting peak values in stages, and ensuring the safety of the fan through self-adaptive linkage.

Inventors

  • LIU YAN
  • HAO RONG

Assignees

  • 山西职业技术学院

Dates

Publication Date
20260505
Application Date
20260122

Claims (10)

  1. 1. The method for measuring, collecting, analyzing and alarming the swing amplitude of the tower drum of the fan in real time is characterized by comprising the following steps: s1, synchronously acquiring tower swing data, mooring system tension data and typhoon environment characteristic data, identifying instantaneous loosening and tightening states of a mooring rope, constructing a mooring tension-tower swing dynamic coupling model adapting to the typhoon full stage, and performing decoupling treatment to obtain tower pure swing data; s2, constructing a staged multi-parameter mapping relation prediction model based on a typhoon full-stage load database, and predicting tower barrel swing peaks and variation trends of different stages of typhoons; S3, setting a swing amplitude-mooring tension-instantaneous state collaborative alarm threshold matched with a typhoon stage, and when multiple parameters meet alarm conditions, starting an adaptive linkage intervention mechanism of an active yaw and tower damper of the engine room, and dynamically adjusting an intervention strategy to form a fully closed loop treatment flow of 'state identification-monitoring-stage prediction-adaptive intervention'.
  2. 2. The method for measuring, collecting, analyzing and alarming the swing amplitude of the tower drum of the fan in real time according to claim 1 is characterized in that in the S1, multi-dimensional inertial sensing units are arranged at the top, the middle and the bottom of the tower drum, tension sensing units are arranged at the top and the middle of a mooring cable, and typhoon environment sensing units are arranged at the outer side of a cabin; each sensing unit synchronously starts data acquisition based on a unified time reference, and respectively acquires the swing angle and angular speed data of the tower, the real-time tension data of the mooring system and the wind direction and wind speed characteristic data of typhoon environment; Preprocessing the collected original data, eliminating the interference of environmental noise, and obtaining standardized multi-source perception data.
  3. 3. The method for measuring, collecting, analyzing and alarming the swing amplitude of the tower drum of the blower according to claim 1, wherein in the step S1, based on the collected tension data of the mooring system, the tension mutation characteristic and the duration parameter are extracted, and the instantaneous loosening state and the instantaneous tightening state of the mooring cable are identified; Dividing three stages before typhoon logging, during logging and after logging according to typhoon environment characteristic data, combining mooring cable state identification results and tower swing data of each stage, extracting relevant characteristic parameters of mooring tension and tower swing under different stages, and constructing a dynamic coupling model comprising stage adaptation items and mooring state correction items; performing decoupling operation on multi-source data in the dynamic coupling model by adopting an improved self-adaptive Kalman filtering algorithm, and removing interference of dynamic changes of a mooring system on swing data of the tower drum in different wind stages in a targeted manner to finally obtain pure swing data reflecting the true stress state of the tower drum; in order to realize the dynamic coupling model construction and decoupling operation, the following algorithm formula is adopted: ; Wherein, the Is that The tower drum is coupled with the swinging angle at the moment, Is the oscillation coefficient of the foundation of the tower, Is that The moment tower barrel theory basic swinging angle, For the mooring tension-angle conversion factor, Is that The real-time tension of the mooring system is realized at the moment, For the mooring tension reference value, For the mooring state angle correction factor, Is that The status of the mooring line is identified at the moment, Is that A moment environmental interference error term; improved adaptive Kalman filtering decoupling algorithm: firstly, carrying out state prediction and prediction error covariance update: ; ; and then carrying out Kalman gain calculation and state updating: ; ; Wherein, the Is that The pure swing state estimated value of the tower barrel after time decoupling, Is that Time of day based The state prediction value of the moment in time, In the form of a state transition matrix, In order for the system to be process noisy, Is that The time instant prediction error covariance matrix, Is the transpose of the state transition matrix, In order to process the noise covariance matrix, Is that The time of day kalman gain is calculated, Is that Coupling data observations acquired by time-of-day multisource sensing, In order to observe the matrix, For the transposition of the observation matrix, Is that And observing a noise covariance matrix at the moment.
  4. 4. The method for measuring, collecting, analyzing and alarming the swing amplitude of the tower drum of the fan in real time according to claim 2, wherein the multidimensional inertial sensing unit and the tension sensing unit are both in a redundant packaging design for resisting strong wind impact, and a data collecting load monitoring module and a self-adaptive sampling adjusting module are arranged in the redundant packaging design; When abnormal data acquisition load caused by typhoon extreme load is monitored, the self-adaptive sampling adjustment module automatically adjusts a sampling strategy according to the current typhoon stage and the state of the mooring rope.
  5. 5. The method for real-time measurement, acquisition analysis and alarm processing of swing amplitude of a tower of a fan according to claim 1, wherein in S2, the construction of the staged multi-parameter mapping relation prediction model specifically comprises: Collecting wind direction and wind speed data of typhoons in all stages in typhoons in different sea areas and different strength, mooring tension data, tower swing data and fan structure safety feedback data, and constructing a standardized typhoons all-stage load database; Classifying and labeling database data according to three stages before and during typhoon logging in; respectively taking the typhoon environment characteristic data and mooring state data normalized in each stage as input parameters and the swing amplitude of the tower drum as output parameters to construct a staged multi-parameter mapping relation prediction model; Training and optimizing the prediction model of each stage through historical data in a database; For constructing the staged multi-parameter mapping relation prediction model, an improved BP neural network model is adopted, and the mathematical expression is as follows: ; Wherein, the Is that The output value of the time instant prediction model, In order for the hidden layer to activate a function, For the output layer to activate a function, To input layer no From neuron to hidden layer The weight of the individual neurons is determined, To hidden layer (L) Individual neurons to output layer The weight of the individual neurons is determined, To hidden layer (L) The bias term of the individual neurons is set, To the output layer The bias term of the individual neurons is set, Is that Normalized input parameter vector of the time of day model, For the number of neurons in the hidden layer, Is the number of neurons of the input layer.
  6. 6. The method for real-time measurement, acquisition analysis and alarm processing of swing amplitude of a tower of a fan according to claim 5, wherein in S2, the method for predicting the swing peak value and the change trend of the tower in different stages of typhoons specifically comprises: collecting environmental characteristic data of the current typhoon in real time, and judging the current typhoon stage; Inputting typhoon environment characteristic data of the current stage and synchronously acquired mooring state data into a multi-parameter mapping relation prediction model of the corresponding stage to obtain tower swing amplitude variation trends of the current stage and the subsequent stage; extracting swing peak value data of each stage and the pre-judging time of the occurrence of the peak value from the change trend, and taking the swing peak value data and the pre-judging time as core basis for the establishment of subsequent alarm judgment and intervention strategies; Swing amplitude predicted value output based on prediction model The mathematical expression of the oscillation peak values of different stages of typhoons is extracted as follows, wherein the typhoon stage time interval is divided by typhoon environment characteristic data: ; ; Wherein, the Is the first The tower drum swing peak value of each typhoon stage, The sign is calculated for the maximum value, Is that The predicted value of the moment swing amplitude, As a function of the time variable, Is the first The time intervals of the individual typhoon phases, Is the first The pre-judging time of the occurrence of the swing peak value in the typhoon stage; To obtain the sign of the argument corresponding to the maximum value, Identified for typhoon phases.
  7. 7. The method for real-time measurement, acquisition analysis and alarm processing of swing amplitude of a wind turbine tower according to claim 1, wherein in the step S3, based on the structural safety design requirement of a deep sea floating wind turbine, corresponding swing amplitude reference threshold and mooring tension reference threshold are respectively set for three stages before, during and after typhoon landing by combining structural safety feedback data in a typhoon full-stage load database; setting dynamic correction coefficients for reference thresholds of each stage aiming at the instantaneous loosening and tightening states of the mooring rope to form a cooperative alarm threshold system which is matched with the typhoon stage and the mooring state; when the real-time monitoring tower pure swing data reach swing amplitude reference threshold values of corresponding stages, and the synchronously monitored mooring tension data reach the mooring tension reference threshold values of the corresponding stages, or the mooring rope is monitored to have instantaneous loosening/tightening states and accompanied with swing data mutation, the early warning of the corresponding grade is judged to be triggered.
  8. 8. The method for measuring, collecting, analyzing and alarming the swing amplitude of the tower drum of the fan in real time according to claim 7, wherein in the step S3, after the early warning is triggered, an initial linkage intervention strategy is formulated according to the typhoon stage, the mooring rope state and the pre-judged swing peak value data, and a cooperative control signal is sent to a yaw control system of the engine room and a damper control system of the tower drum; the cabin yaw control system adjusts the cabin direction to an optimal wave avoiding angle according to typhoon wind direction data, and reduces the forward impact force of typhoon on the tower drum; The tower damper control system starts damping adjustment modes of corresponding grades according to the tower pure swing data to inhibit the expansion of the swing amplitude of the tower; in the intervention process, tower swing data and mooring tension data are collected in real time, the intervention effect is dynamically evaluated, and yaw angle and damping adjustment parameters are continuously optimized; The calculation formula of the yaw angle and damping adjustment of the engine room is as follows: ; ; Wherein, the Is that The moment the nacelle is at an optimal yaw angle, For the initial yaw angle of the wind turbine, For a yaw adjustment factor that is correlated to the yaw angle, Is that The pure swing angle of the tower drum is changed at any moment, As the swing peak of the current typhoon stage, For a yaw adjustment coefficient associated with the mooring tension, Is that The real-time tension of the mooring system is realized at the moment, For the mooring tension reference threshold for the current typhoon phase, Is that The optimal damping coefficient of the tower damper at any moment, As a result of the initial damping coefficient, For the damping adjustment coefficient associated with the oscillation angle, For a damping adjustment coefficient associated with the moored state, Is that And (5) marking the state of the mooring rope at the moment.
  9. 9. The method for measuring, collecting, analyzing and processing the swing amplitude of the tower drum of the fan in real time according to claim 1, further comprising the steps of multi-mode local warning and risk classification prompting: When the cooperative alarm threshold is triggered, corresponding multi-mode local alarms are started according to the early warning level, wherein the multi-mode local alarms comprise an audible and visual alarm unit arranged in a cabin, an indicator light alarm unit of a floating platform and a voice prompt unit of an operation and maintenance cabin; Meanwhile, risk prompt information is generated, and the current typhoon stage, mooring rope state, abnormal swing data condition and intervention implementation state are clearly marked.
  10. 10. The method for measuring, collecting, analyzing and processing the swing amplitude of the tower drum of the fan in real time according to claim 2, further comprising the steps of fault-tolerant switching and fault tracing of the multi-source sensing unit: Monitoring the working states and the data transmission states of the multi-dimensional inertial sensing units, the tension sensing units and the typhoon environment sensing units in real time; when a certain sensing unit is detected to have a fault or abnormal data transmission, the sensing unit is automatically switched to a preset redundant sensing unit.

Description

Real-time measuring, collecting, analyzing and alarming processing method for swing amplitude of fan tower Technical Field The invention relates to the technical field of wind power monitoring, in particular to a method for measuring, collecting, analyzing and alarming the swing amplitude of a fan tower in real time. Background The deep-open sea floating type wind turbine is used as core equipment for ocean renewable energy development, and the working environment of the deep-open sea floating type wind turbine is subjected to the examination of extreme meteorological hydrologic conditions such as typhoons, strong waves and the like all the year round, wherein the coupling effect of strong wind load and dynamic response of a mooring system in the whole stage of typhoons (before, during and after landing) becomes a key factor for threatening the structural safety of the wind turbine. In the typhoon influence process, the mooring rope can frequently switch instantaneous loosening and tightening states along with the dynamic changes of wind speed and wave height, and the state changes and the swing of the tower barrel form strong coupling interference. In the existing measurement method, only the tower swing data or the mooring tension data are singly collected, the dynamic coupling relation between the two is not considered, the interference caused by dynamic change of a mooring system cannot be effectively eliminated, the obtained tower swing data contain a large amount of redundancy errors, and the real stress state of the tower is difficult to reflect. Based on the distortion data, the existing prediction model cannot accurately predict tower swing peak values and change trends of different stages of typhoons, and the alarm threshold value is not adapted to dynamic characteristics of typhoons and mooring states, so that the intervention strategy is lack of pertinence, early warning hysteresis or excessive/insufficient intervention often occurs in the typhoons strong impact stage, and the structural integrity and operation safety of the deep-open sea floating fan are seriously threatened. Disclosure of Invention The invention aims to provide a method for measuring, collecting, analyzing and alarming the swing amplitude of a tower drum of a fan in real time, so as to solve the technical problems of data distortion, inaccurate prediction alarming and lack of pertinence in intervention caused by mooring and tower drum swing coupling in the existing typhoon environment. In order to solve the technical problems, the invention provides a method for measuring, collecting, analyzing and alarming the swing amplitude of a fan tower in real time, which comprises the following steps: s1, synchronously acquiring tower swing data, mooring system tension data and typhoon environment characteristic data, identifying instantaneous loosening and tightening states of a mooring rope, constructing a mooring tension-tower swing dynamic coupling model adapting to the typhoon full stage, and performing decoupling treatment to obtain tower pure swing data; s2, constructing a staged multi-parameter mapping relation prediction model based on a typhoon full-stage load database, and predicting tower barrel swing peaks and variation trends of different stages of typhoons; S3, setting a swing amplitude-mooring tension-instantaneous state collaborative alarm threshold matched with a typhoon stage, and when multiple parameters meet alarm conditions, starting an adaptive linkage intervention mechanism of an active yaw and tower damper of the engine room, and dynamically adjusting an intervention strategy to form a fully closed loop treatment flow of 'state identification-monitoring-stage prediction-adaptive intervention'. Preferably, in the step S1, multidimensional inertial sensing units are arranged at the top, middle and bottom of the tower, tension sensing units are arranged at the top and middle sections of the mooring cable, and typhoon environment sensing units are arranged at the outer side of the nacelle; each sensing unit synchronously starts data acquisition based on a unified time reference, and respectively acquires the swing angle and angular speed data of the tower, the real-time tension data of the mooring system and the wind direction and wind speed characteristic data of typhoon environment; Preprocessing the collected original data, eliminating the interference of environmental noise, and obtaining standardized multi-source perception data. Preferably, in the step S1, based on the collected tension data of the mooring system, extracting a tension abrupt change feature and a duration parameter, and identifying an instantaneous slack state and an instantaneous tight state of the mooring line; Dividing three stages before typhoon logging, during logging and after logging according to typhoon environment characteristic data, combining mooring cable state identification results and tower swing data of each stage, extracting re