CN-120926025-B - Wind power generation tower control method and system based on fatigue damage driving

CN120926025BCN 120926025 BCN120926025 BCN 120926025BCN-120926025-B

Abstract

The invention discloses a wind power generation tower control method and system based on fatigue damage driving, and relates to the technical field of wind power generation control, comprising the following steps of S1, acquiring tower barrel structure response data, three-dimensional wind field parameters and power grid dispatching instructions of all wind power generation towers in a wind power field in real time through a distributed sensing network; step S2, constructing a digital twin model of the tower structure, dynamically predicting the residual fatigue life and damage evolution trend of key parts of each tower based on real-time structural response data, step S3, executing double-layer collaborative optimization control, step S4, issuing an optimization instruction to an executing mechanism, and updating the digital twin model through real-time feedback data. The wind power generation tower control method and system based on fatigue damage driving realize safe operation guarantee, fatigue damage inhibition and power generation efficiency collaborative optimization of the wind power generation tower, and balance the service life and power generation performance of the unit.

Inventors

LUO YUXIAO
HAN RUI
CAI QINLIN
Dai kaoshan

Assignees

四川大学

Dates

Publication Date: 20260508
Application Date: 20250818

Claims (6)

1. The wind power generation tower control method based on fatigue damage driving is characterized by comprising the following steps of: step S1, acquiring tower barrel structure response data, three-dimensional wind field parameters and power grid dispatching instructions of all wind power generation towers in a wind power plant in real time through a distributed sensing network; s2, constructing a digital twin model of the tower structure, and dynamically predicting the residual fatigue life and damage evolution trend of key parts of each tower based on real-time structural response data; s3, executing double-layer collaborative optimization control; The local control layer is used for generating a variable pitch and damper adjusting instruction by adopting a self-adaptive model predictive control algorithm according to a vibration mode predictive result output by the tower digital twin model and inhibiting tower resonance; the central coordination layer dynamically adjusts the power distribution and yaw angle of the unit by taking the residual fatigue life of each tower as a constraint condition, so that the offset of the wake center line relative to the incoming wind direction is more than or equal to 15 degrees to avoid a high fatigue risk area; in step S3, the adaptive model predictive control algorithm includes predicting an overhead displacement response spectrum within 30 seconds based on the digital twin model, and solving an objective function to generate an optimal pitch sequence, as shown in the following formula: ; Wherein, the Real-time vibration acceleration; The variable quantity of the pitch angle at the kth moment; The vibration suppression weight coefficient is; Punishment weights for pitch motions, N being a prediction time domain; In step S3, the optimization function of the central coordination layer is defined as follows: ; Wherein, P i is the real-time power of the ith wind power generator set, D j is the fatigue damage increment of the high risk set j; H is a set of high-risk units; The high risk unit set H is represented by the following formula: ; numbering the wind generating set; The remaining life percentage of the unit k; The arithmetic average value of the residual life of the whole-field unit; in step S3, the wake path offset is calculated by solving the upstream train yaw compensation angle The realization is as follows: ; Wherein, the The safe offset distance is calculated according to the position of the high-risk tower barrel, wherein L is the unit distance; in step S3, the central coordination layer implements a tower life balancing strategy when the tower has residual life When it is at its power set point Adjustment is performed as shown in the following formula: ; Wherein, the Rated power of the unit; The residual life percentage of the unit; at the same time, cut load is transferred to >1.2 Is a unit of (a); and S4, issuing an optimization instruction to the executing mechanism, and updating the digital twin model through real-time feedback data.
2. The wind power generation tower control method based on fatigue damage driving according to claim 1 is characterized in that in the step S1, the distributed sensing network comprises a high-density fiber grating sensor array, a pulse laser radar, a temperature and humidity, vibration and lightning stroke monitoring unit, the high-density fiber grating sensor array is arranged on the inner wall of a tower barrel, the high-density fiber grating sensor array is arranged in a group every 90 degrees along the circumferential direction of the tower barrel, the pulse laser radar is arranged at the front edge and a wake area of a wind power plant, and the temperature and humidity, vibration and lightning stroke monitoring unit is arranged at the top of the tower.
3. The method according to claim 2, wherein in step S1, the data sampling frequency of the fiber bragg grating sensor array is dynamically adjusted.
4. The method for controlling a wind power generation tower based on fatigue damage driving according to claim 1, wherein in step S3, the stress change rate of the key measurement point is calculated in real time as 。
5. The method according to claim 1, wherein in step S2, the digital twin model comprises a parameterized structural mechanics model based on finite element analysis, a fatigue life predictor trained by LSTM-transducer neural network, and a parameter self-correction module receiving control verification data.
6. The wind power generation tower control system based on fatigue damage driving according to claim 1 is characterized by comprising a sensing layer, an edge calculation layer, a central decision layer and an execution layer, wherein a real-time data bus with end-to-end time delay is connected with each layer, the sensor comprises a fiber grating sensor array, a pulse laser radar group and a hardware network of an environment monitoring unit, the edge calculation layer comprises a local controller deployed on each tower barrel and is configured with a dynamic adaptation processor to execute a vibration suppression algorithm, the central decision layer comprises a space-time database engine and a server of an optimization solver, and the execution layer comprises a pitch system, an active damper and a yaw driver.

Description

Wind power generation tower control method and system based on fatigue damage driving Technical Field The invention relates to the technical field of wind power generation control, in particular to a wind power generation tower control method and system based on fatigue damage driving. Background The traditional control strategy aims at maximizing the power of a single machine, ignores the fatigue accumulation effect of the tower, and increases the fracture risk in the service period. The transient resonance cannot be captured only by monitoring with a limited strain gauge with a sampling rate of less than 1 Hz. The wake flow influence among units in the wind power plant causes that the power generation loss is only passively avoided in the existing yaw control, the structural state of a downstream tower barrel is not considered, the structural damage diagnosis is decoupled from the power control, the control instruction generation delay exceeds 60 seconds, and the power grid frequency modulation requirement is difficult to adapt. Disclosure of Invention The invention aims to provide a wind power generation tower control method and system based on fatigue damage drive, which aim at solving the problems of high tower barrel fatigue accumulation risk, insufficient transient resonance monitoring, insufficient wake flow influence avoidance, delayed control response and the like in the traditional wind power generation tower control, and realize the safe operation guarantee, fatigue damage inhibition and power generation efficiency collaborative optimization of a wind power generation tower through a multisource sensing and dynamic optimization technology, so that the service life and the power generation performance of a unit are balanced. The invention provides a wind power generation tower control method and system based on fatigue damage driving, comprising the following steps of S1, acquiring tower barrel structure response data, three-dimensional wind field parameters and power grid dispatching instructions of each wind power generation tower in a wind power field in real time through a distributed sensing network; s2, constructing a digital twin model of the tower structure, and dynamically predicting the residual fatigue life and damage evolution trend of key parts of each tower based on real-time structural response data; s3, executing double-layer collaborative optimization control; The local control layer is used for generating a variable pitch and damper adjusting instruction by adopting a self-adaptive model predictive control algorithm according to a vibration mode predictive result output by the tower digital twin model and inhibiting tower resonance; the central coordination layer dynamically adjusts the power distribution and yaw angle of the unit by taking the residual fatigue life of each tower as a constraint condition, so that the offset of the wake center line relative to the incoming wind direction is more than or equal to 15 degrees to avoid a high fatigue risk area; and S4, issuing an optimization instruction to the executing mechanism, and updating the digital twin model through real-time feedback data. Preferably, in step S1, the distributed sensing network includes a high-density fiber bragg grating sensor array, a pulse laser radar, a temperature, humidity, vibration and lightning stroke monitoring unit, the high-density fiber bragg grating sensor array is arranged on the inner wall of the tower, the high-density fiber bragg grating sensor array is arranged in a group every 90 degrees along the circumferential direction of the tower, the pulse laser radar is deployed at the front edge and wake area of the wind farm, and the temperature, humidity, vibration and lightning stroke monitoring unit is arranged at the top of the tower. Preferably, in step S1, the data sampling frequency of the fiber bragg grating sensor array is dynamically adjusted. Preferably, in step S3, the adaptive model predictive control algorithm includes predicting an overhead displacement response spectrum within 30 seconds based on the digital twin model, and solving an objective function to generate an optimal pitch sequence, as shown in the following formula: wherein a (t) is real-time vibration acceleration, delta theta (k) is variable quantity of a variable pitch angle at a kth moment, omega 1 is a vibration suppression weight coefficient, omega 2 is a variable pitch action punishment weight, and N is a prediction time domain. Preferably, in step S3, the optimization function of the central coordination layer is defined as follows: Wherein P i is the real-time power of the ith wind power generator unit, D j is the fatigue damage increment of the high-risk unit j, lambda is the damage penalty coefficient, and H is the high-risk unit set; The high risk unit set H is represented by the following formula: k is the number of the wind generating set; The remaining life percentage of the unit k; the arithmetic ave