CN-121993345-A - Wave pitching motion control method and system for offshore floating wind turbine generator

CN121993345ACN 121993345 ACN121993345 ACN 121993345ACN-121993345-A

Abstract

The invention discloses a wave pitching motion control method and system of an offshore floating wind turbine, and belongs to the technical field of offshore wind power equipment control. The method comprises the steps of signal acquisition, real-time acquisition of unit running state, marine environment and pitching motion state parameters, wave load prediction, future wave load prediction based on an improved LSTM neural network model, pitching motion state assessment, motion trend pre-judgment and judgment whether a safety threshold is exceeded or not through a dynamic model, cooperative control instruction generation, control execution and feedback correction, actuation of an actuating mechanism and dynamic correction of control instructions, wherein the cooperative control instruction generation is based on an MPC algorithm and comprises the steps of generating a variable pitch angle, generator torque and cabin yaw angle fine adjustment. According to the invention, through accurately predicting wave load, cooperatively adjusting multiple controllable components and dynamically correcting, efficient and accurate inhibition of pitching motion is realized, and the running stability, wind energy capturing efficiency and equipment safety of the unit are improved.

Inventors

ZHANG JIANHUA
SUN TIANXUE
SUN KE
ZHENG YU
ZHONG LEI
WEI MINGYANG
LIU ZHICHENG
YUE FENG

Assignees

哈尔滨工程大学

Dates

Publication Date: 20260508
Application Date: 20260123

Claims (10)

1. The wave pitching motion control method of the offshore floating wind turbine generator is characterized by comprising the following steps of: S1, collecting signals, namely collecting running state parameters, marine environment parameters and pitching motion state parameters of a floating wind turbine generator in real time through a sensor group arranged on the floating wind turbine generator; s2, predicting wave load, namely predicting wave load acting on the wind turbine foundation in a preset time period in the future by adopting an improved long-short-term memory LSTM neural network model based on the marine environment parameters acquired in the step S1; S3, evaluating the pitching motion state, namely establishing a pitching motion dynamics model based on the pitching motion state parameters acquired in the step S1 and the wave load predicted in the step S2, evaluating future pitching motion trend of the unit, and judging whether the current pitching motion exceeds a preset safety threshold value; S4, generating a cooperative control instruction, namely if the step S3 judges that the pitching motion exceeds a preset safety threshold, controlling an MPC algorithm based on model prediction, taking the inhibition of the pitching motion and the guarantee of wind energy capturing efficiency as optimization targets, and generating a pitch angle adjusting instruction, a generator torque adjusting instruction and a cabin yaw angle fine adjusting instruction; And S5, controlling execution and feedback correction, namely sending the control instruction generated in the step S4 to a corresponding executing mechanism, driving the executing mechanism to act, and collecting feedback parameters of pitching motion of the unit in real time through a sensor group to dynamically correct the control instruction.
2. The control method according to claim 1, wherein in step S1, the operation state parameters include wind wheel rotation speed, generator power, pitch angle current value, nacelle yaw current value, the marine environment parameters include effective wave height, wave period, wave direction, wind speed, wind direction, the pitch motion state parameters include pitch angle, pitch angle speed, pitch angle acceleration, and the sensor group includes a wave height meter, an anemometer, a gyroscope, a rotation speed sensor, a power sensor, and an angle sensor.
3. The control method according to claim 1, wherein in step S2, the improved LSTM neural network model is constructed and trained as follows: S21, constructing a data set, namely collecting historical marine environment parameters and corresponding wave load data, carrying out normalization processing on the data, and dividing a training set, a verification set and a test set; S22, designing a model structure, wherein the improved LSTM neural network model comprises an input layer, a hidden layer, an attention mechanism layer and an output layer, wherein the input layer receives normalized ocean environment parameters, the hidden layer adopts a double-layer LSTM structure, the attention mechanism layer is used for enhancing weight distribution on key environment parameters, and the output layer outputs wave load predicted values in a future preset time period; S23, training and optimizing the model by using a training set, adjusting the super parameters of the model by using a verification set, using a mean square error MSE as a loss function, and estimating an Adam optimization algorithm by using a self-adaptive moment to minimize the loss function to obtain an improved LSTM neural network model after training, wherein the loss function is shown in a formula (1): Where N is the number of samples, For the actual wave load value of the i-th sample, Is the predicted value of the wave load for the ith sample.
4. The control method according to claim 1, wherein in step S3, the pitching motion dynamics model is established based on newton' S second law, and the expression is as shown in formula (2): wherein J is the moment of inertia of the wind turbine generator around the pitching axis, Is pitch angle acceleration, C is pitch motion damping coefficient, For pitch angle rate, K for the coefficient of restitution stiffness, Is the pitch angle, t is the time, For the pitching moment generated by the wind load, For the pitching moment generated by the wave load, Reverse pitching moment generated for the control command; Obtaining a pitching motion trend in a future preset time period through numerical integration based on a formula (2), and comparing a predicted pitching angle maximum value with a preset safety threshold value And comparing, and judging whether the pitching motion exceeds the safety range.
5. The control method according to claim 1, wherein in step S4, the process of generating the cooperative control command based on the model predictive control MPC algorithm is as follows: s41, determining a prediction time domain And control time domain And (2) and ; S42, establishing a discretized linear prediction model, as shown in a formula (3): where k is the discrete time step, Is the state vector at time k, Is the control input vector at time k, The method is characterized in that the method is an output vector at the moment k, A is a state matrix, B is an input matrix, C is an output matrix, and D is a direct transmission matrix; s43, determining an optimization objective function, as shown in a formula (4): Wherein, the 、、 Respectively the reference values of pitch angle, pitch angle speed and wind wheel rotating speed, 、、 The control increment of the pitch angle, the generator torque and the cabin yaw angle are respectively, Is a weight coefficient; S44, solving the minimum value of the optimization objective function through a quadratic programming algorithm on the premise of meeting the control input constraint and the output constraint to obtain an optimal control input sequence; S45, extracting a first element of the optimal control input sequence as a control instruction at the current moment.
6. The control method according to claim 1, wherein in step S5, the dynamic correction process is to collect the pitch angle after control in real time And pitch rate Calculating deviation 、 And if the absolute value of the deviation is larger than a preset correction threshold, correcting the control command based on a proportional-integral (PI) controller, wherein the corrected control command is shown in a formula (5): Wherein, the In order to correct the control command after it has been made, Is a coefficient of proportionality and is used for the control of the power supply, As an integral coefficient of the power supply, Is a bias vector.
7. A wave pitch motion control system for an offshore floating wind turbine, configured to implement the control method of any one of claims 1-6, comprising: the signal acquisition module consists of a plurality of sensors and is used for acquiring running state parameters, marine environment parameters and pitching motion state parameters of the unit in real time and transmitting the running state parameters, the marine environment parameters and the pitching motion state parameters to the data processing module; The data processing module is in communication connection with the signal acquisition module and is used for preprocessing acquired parameters and respectively transmitting the preprocessed data to the wave load prediction module and the pitching motion state evaluation module; the wave load prediction module is in communication connection with the data processing module, is internally provided with an improved LSTM neural network model, and is used for predicting the wave load in a preset time period in the future and transmitting the wave load to the pitching motion state evaluation module; The pitching motion state evaluation module is respectively connected with the data processing module and the wave load prediction module in a communication way, is internally provided with a pitching motion dynamic model and is used for evaluating future pitching motion trend of the unit and judging whether the future pitching motion trend exceeds a preset safety threshold value or not; The cooperative control decision module is in communication connection with the pitching motion state evaluation module, is internally provided with a model predictive control MPC algorithm, is used for generating a pitch angle adjustment instruction, a generator torque adjustment instruction and a cabin yaw angle fine adjustment instruction after receiving a control trigger signal, and transmits the pitch angle adjustment instruction, the generator torque adjustment instruction and the cabin yaw angle fine adjustment instruction to the control execution module; The control execution module is in communication connection with the cooperative control decision module and comprises a pitch control execution mechanism, a generator torque adjustment mechanism and a cabin yaw execution mechanism, and is used for receiving control instructions and driving corresponding parts to act; And the feedback correction module is respectively in communication connection with the signal acquisition module and the cooperative control decision module and is used for calculating deviation and dynamically correcting the control instruction based on the controlled pitching motion feedback parameters.
8. The control system of claim 7, wherein the sensors in the signal acquisition module include a wave height gauge deployed near the crew foundation, an anemometer deployed at the top of the nacelle, a gyroscope deployed at the bottom of the tower, a rotational speed sensor deployed on the wind turbine shaft, a power sensor deployed on the generator, and an angle sensor deployed on the pitch and yaw systems.
9. The control system of claim 7, wherein the preprocessing of the data processing module includes filtering denoising using a kalman filter algorithm, normalizing using a min-max normalization method, and the normalization formula is shown in formula (6): Wherein, the For normalized parameter values, x is the original parameter value, As a minimum value of the parameter(s), Is the maximum value of the parameter.
10. The control system of claim 7, further comprising a fault diagnosis module in communication with the signal acquisition module and the control execution module for monitoring the operating conditions of the sensor and the execution mechanism in real time, and when an abnormality occurs, sending out a fault alarm signal and switching to an emergency control mode.

Description

Wave pitching motion control method and system for offshore floating wind turbine generator Technical Field The invention relates to the technical field of offshore wind power equipment control, in particular to a wave pitching motion control method and system of an offshore floating wind turbine generator. Background With the acceleration of the global energy structure transformation process, offshore wind power is taken as an important component of clean renewable energy, and development and utilization of the offshore wind power are widely focused. Compared with a fixed offshore wind turbine, the floating wind turbine can be deployed in deeper sea areas, has the advantages of richer wind energy resources, smaller influence on offshore ecological environment and the like, and is an important direction of future offshore wind power development. However, the foundation (such as semi-submersible, spar, tension leg, etc.) of the floating wind turbine floats on the sea surface, and is susceptible to the action of ocean environmental loads such as waves, wind, ocean currents, etc. to generate motion with multiple degrees of freedom, wherein pitching motion (rotation around the transverse axis of the turbine) has a significant influence on safe and stable operation of the turbine. On one hand, the central position of the wind wheel is periodically lifted due to severe pitching movement, the optimal alignment state of the wind wheel and incoming wind is damaged, and the wind energy capturing efficiency is reduced, on the other hand, the fatigue damage of key components such as a tower, blades and a transmission system of a unit is aggravated by inertial load generated by pitching movement, the service life of equipment is shortened, and even safety accidents such as structural failure and the like are possibly caused. The existing control means for the pitching motion of the floating wind turbine generator mainly comprises passive control and active control. The passive control suppresses pitching motion by optimizing a basic structure design (such as adding ballasts, arranging damping plates and the like) or installing a passive damping device, but has limited control effect and is difficult to adapt to complex and changeable marine environments, and the active control generates reverse moment by actively adjusting controllable components (such as pitch angle, cabin yaw angle, generator torque and the like) of a unit so as to counteract pitching motion caused by wave load, so that the passive control is a hot spot direction of current research. However, the existing active control method still has many defects that firstly, the algorithm is mostly based on a traditional PID control algorithm, the adaptability of the algorithm to a nonlinear and time-varying floating wind turbine generator system is poor, the control accuracy is low under the action of complex wave load (such as irregular waves), secondly, the control strategy is used for singly adjusting the pitch angle or the generator torque, the cooperative action among controllable components is not fully considered, the control efficiency is low, thirdly, the wave load is not accurately predicted, the control is only carried out by relying on real-time feedback signals, and control lag exists, so that high-frequency or sudden pitching motion is difficult to effectively inhibit. Therefore, the invention provides the wave pitching motion control method and system for the offshore floating wind turbine generator, which can accurately predict wave load, cooperate with a plurality of controllable components and adapt to complex ocean environments. Disclosure of Invention The invention aims to overcome the defects in the prior art and provides a wave pitching motion control method and system for an offshore floating wind turbine, and the method realizes high-efficiency and accurate suppression of pitching motion by accurately predicting wave load and cooperatively adjusting a plurality of controllable components, so that the running stability and safety of the floating wind turbine are improved. In order to achieve the above purpose, the invention adopts the following technical scheme: a wave pitching motion control method of an offshore floating wind turbine generator comprises the following steps: S1, collecting signals, namely collecting running state parameters, marine environment parameters and pitching motion state parameters of a floating wind turbine generator in real time through a sensor group arranged on the floating wind turbine generator; s2, predicting wave load, namely predicting wave load acting on the wind turbine foundation in a preset time period in the future by adopting an improved long-short-term memory LSTM neural network model based on the marine environment parameters acquired in the step S1; S3, evaluating the pitching motion state, namely establishing a pitching motion dynamics model based on the pitching motion state parameters acquire