Variable pitch control optimization method for offshore wind turbine generator

Abstract

According to the variable pitch control method of the wind turbine generator, a dynamic decision-making mechanism is constructed by fusing a physical model and an LSTM prediction algorithm, vibration suppression and a safety grading execution strategy are integrated, all-working-condition optimization control of the wind turbine generator is achieved, the load peak value is accurately predicted based on the real-time wind speed, thrust and vibration data, and the control accuracy is improved. A cost function with thrust fluctuation and variable pitch gentle double targets is adopted to optimize a pitch angle sequence, hub axial load fluctuation is reduced, vibration energy is effectively reduced by detecting a tower vibration amplitude dynamic superposition anti-phase compensation angle, the control precision error is reduced, and the power generation efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the field of wind power generation, and in particular to a method for optimizing pitch control of an offshore wind turbine. Background Art

[0002] With the development of renewable energy technology, especially the growing demand in the field of offshore wind power, higher requirements are placed on the stability, reliability and efficiency of large offshore wind turbines in complex environments.

[0003] Existing pitch control systems often struggle to simultaneously meet the demands of multi-objective coordinated control, including power regulation, load optimization, and safety protection. Especially at wind speeds above rated, aerodynamic imbalances significantly increase blade flapping loads and tower overturning loads, posing challenges to the safe operation of turbines. Summary of the Invention

[0004] In a first aspect of the present disclosure, a method for controlling a pitch of a wind turbine is provided, comprising the following steps:

[0005] Acquire first operating data of the wind turbine generator set in a first time period, where the first operating data includes wind speed time series data, hub axial thrust, pitch angle, and tower vibration spectrum of the wind turbine generator set in the first time period, where the end time of the first time period is real time;

[0006] The first operation data is input into a pre-trained prediction model to calculate a target instruction, wherein the training process of the prediction model includes:

[0007] Acquire second operating data of a second time period of the wind turbine generator set, where the end time of the second time period is the start time of the first time period,

[0008] predicting the thrust peak value of the first time period according to the second operating data, and calculating a first target instruction based on the thrust peak value, wherein the first target instruction includes a pitch angle sequence,

[0009] Based on the first target instruction, a second target instruction is calculated in combination with the amplitude constraint and the rate constraint of the blade, the aerodynamic load of the blade under the second target instruction is calculated, and the training is terminated when the aerodynamic load meets a set threshold;

[0010] The wind turbine generator set is controlled to change pitch according to the target instruction.

[0011] In combination with the first aspect, the predicted thrust peak value in the first time period is calculated using the following formula:

[0012]

[0013] Where ρ is the air density, R is the radius of the wind wheel, Cp is the wind energy utilization coefficient, λ is the tip speed ratio, β is the pitch angle, V w is the wind speed.

[0014] In combination with the first aspect, the calculating the first target instruction based on the thrust peak includes calculating the optimal pitch angle when the predicted thrust peak exceeds a design safety threshold, and otherwise maintaining the current pitch angle, specifically including:

[0015] Establish a cost function with thrust fluctuation coefficient and pitch smoothness coefficient as optimization objectives;

[0016] The minimum value of the cost function is solved by quadratic programming to obtain the optimal pitch angle.

[0017] In combination with the first aspect, the second target instruction calculated based on the first target instruction and combined with the amplitude constraint and rate constraint of the blade includes screening the instructions whose pitch amplitude is within the amplitude constraint range and the pitch speed is within the rate constraint range as the second target instruction.

[0018] In combination with the first aspect, the second operating data includes historical wind speed data, pitch angle execution sequence, actual thrust measurement value and tower vibration spectrum.

[0019] In combination with the first aspect, the training process uses a long short-term memory network to model the relationship between thrust and vibration, the input layer includes wind speed, pitch angle, and tip speed ratio, and the output layer includes thrust prediction values ​​and tower vibration frequency amplitude.

[0020] In combination with the first aspect, the method further includes adding vibration suppression after calculating the target instruction, and when the frequency amplitude in the tower vibration spectrum exceeds a preset threshold, superimposing an anti-phase compensation angle in the pitch angle sequence.

[0021] In combination with the first aspect, before controlling the wind turbine to change the pitch according to the target instruction, a safety verification is performed to compare the difference between the pitch angle output by the prediction model and the current actual pitch angle.

[0022] According to a second aspect of the present disclosure, an electronic device is provided, comprising:

[0023] one or more processors;

[0024] The storage unit is used to store one or more programs, which, when executed by the one or more processors, enable the one or more processors to implement the pitch control method of the wind turbine generator set.

[0025] According to a third aspect of the present disclosure, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the pitch control method of the wind turbine generator system can be implemented.

[0026] Beneficial effects: The present invention provides a pitch control method for a wind turbine, which constructs a dynamic decision-making mechanism by integrating a physical model with an LSTM prediction algorithm, and integrates vibration suppression and safety graded execution strategies, thereby realizing optimized control of the wind turbine under all working conditions. It accurately predicts the load peak based on real-time wind speed, thrust and vibration data, and optimizes the pitch angle sequence using a cost function with dual objectives of thrust fluctuation and pitch smoothness, thereby reducing the hub axial load fluctuation. By detecting the tower vibration amplitude and dynamically superimposing the anti-phase compensation angle, it effectively reduces vibration energy, reduces control accuracy error, and improves power generation efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 The present invention is a flowchart of a pitch control method for a wind turbine generator system according to an embodiment of the present invention.

[0028] Figure 2 The electronic device disclosed herein. DETAILED DESCRIPTION

[0029] Exemplary embodiments are described in detail herein, with examples illustrated in the accompanying drawings. When the following description refers to the drawings, identical numerals in different drawings represent identical or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments are not intended to represent all embodiments consistent with the present disclosure.

[0030] The terms used in the embodiments of the present disclosure are for the purpose of describing specific embodiments only and are not intended to limit the embodiments of the present disclosure. The singular forms "a," "the," and "the" used in the embodiments of the present disclosure and the appended claims are also intended to include plural forms unless the context clearly indicates otherwise. It should also be understood that the term "and / or" used herein refers to and includes any or all possible combinations of one or more associated listed items.

[0031] It should be understood that although the terms first, second, third, etc. may be used to describe various information in the embodiments of the present disclosure, such information should not be limited to these terms. These terms are only used to distinguish information of the same type from each other. For example, without departing from the scope of the embodiments of the present disclosure, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0032] like Figure 1 FIG. 1 is a flow chart of a pitch control method for a wind turbine according to an embodiment of the present disclosure, comprising:

[0033] S101: Acquire first operating data of a wind turbine generator set in a first time period, where the first operating data includes time series data of wind speed, hub axial thrust, pitch angle, and tower vibration spectrum of the wind turbine generator set in the first time period, and an end time of the first time period is real time;

[0034] For example, the following real-time data is collected synchronously at a sampling frequency of 10 Hz using a lidar on the nacelle roof, a wheel hub stress sensor, and a tower base accelerometer:

[0035] Wind speed time series data: Pre-processing uses sliding average filtering to eliminate gust noise

[0036] Hub axial thrust: Based on strain gauge measurements.

[0037] Pitch angle: read the pitch motor encoder feedback value, accuracy ±0.1°

[0038] Tower vibration spectrum: Perform FFT analysis on the acceleration signal to extract the 1P / 3P characteristic frequency amplitude

[0039] Execution layer: The embedded system encapsulates the above data into time-stamp aligned data packets and transmits them to the main controller via the CAN bus.

[0040] S102: Inputting the first operation data into a pre-trained prediction model to calculate a target instruction, wherein the training process of the prediction model includes:

[0041] Acquire second operating data of a second time period of the wind turbine generator set, where the end time of the second time period is the start time of the first time period,

[0042] predicting the thrust peak value of the first time period according to the second operating data, and calculating a first target instruction based on the thrust peak value, wherein the first target instruction includes a pitch angle sequence,

[0043] Based on the first target instruction, a second target instruction is calculated in combination with the amplitude constraint and the rate constraint of the blade, the aerodynamic load of the blade under the second target instruction is calculated, and the training is terminated when the aerodynamic load meets a set threshold;

[0044] For example, the operating records of similar working conditions in the past 24 hours (second time period) are automatically read (including wind speed, blade angle, thrust and vibration).

[0045] Predict future loads: Based on historical patterns, deduce the maximum hub thrust that may occur in the next 30 seconds (first time period).

[0046] Generate preliminary commands: If the predicted thrust exceeds the safety limit: Calculate the optimal pitch angle sequence (balance between reducing thrust fluctuations and smooth pitch change)

[0047] If the thrust is predicted to be safe: maintain the current pitch angle

[0048] Security fix: Two filters are applied to the initial instructions:

[0049] Limit the single pitch angle to no more than the mechanical limit (e.g. ±8°)

[0050] Limit the pitch speed to not exceed the system's tolerance (e.g. 5° per second)

[0051] Optionally, the tower vibration frequency is monitored in real time: when the vibration of a specific frequency (such as the fundamental frequency of blade rotation) exceeds the standard, a small angle compensation is automatically superimposed on the original instruction (the compensation direction is opposite to the vibration phase).

[0052] S103: Controlling the wind turbine generator set to change pitch according to the target instruction.

[0053] For example, if a large pitch change is required (e.g., more than 10°), execute half of the angle first, monitor the thrust change rate for 3 seconds to confirm that there is no drastic fluctuation, and then execute the remaining angle.

[0054] Small-angle pitch changes are executed directly.

[0055] Physical execution: The pitch motor receives the angle command, drives the blades to rotate to the target angle through the gearbox, and verifies the deviation between the actual angle and the command in real time (an alarm is triggered if the deviation exceeds the target angle).

[0056] Optional, model self-optimization: compares actual thrust with predicted values. When the prediction error is significant for multiple consecutive times, the prediction model is automatically updated with the latest data.

[0057] Furthermore, the predicted thrust peak value in the first time period is calculated by the following formula:

[0058]

[0059] Where ρ is the air density, R is the radius of the wind wheel, C p is the wind energy utilization coefficient, λ is the tip speed ratio, β is the pitch angle, V w is the wind speed.

[0060] For example, the air density ρ is obtained by looking up the data from the cabin temperature and humidity sensor, the rotor radius R is a fixed design parameter (such as 120 meters) pre-stored in the control system, and the wind energy utilization coefficient C is p : Retrieve the aerodynamic performance curve corresponding to the current pitch angle β, and the tip speed ratio λ: Calculated in real time by the wind speed v and the generator speed (λ=ωR / v).

[0061] Execution process: Substitute the above parameters into the formula every 0.5 seconds to calculate the maximum thrust value within the next 10 seconds.

[0062] Further, the calculating of the first target instruction based on the thrust peak includes calculating the optimal pitch angle when the predicted thrust peak exceeds a design safety threshold, and otherwise maintaining the current pitch angle, specifically including:

[0063] Establish a cost function with thrust fluctuation coefficient and pitch smoothness coefficient as optimization objectives;

[0064] The minimum value of the cost function is solved by quadratic programming to obtain the optimal pitch angle.

[0065] For example, the design safety threshold is set to 130% of the rated thrust (such as 3000 kN).

[0066] Optimization calculation when exceeding the standard and construction of cost function:

[0067] Thrust fluctuation coefficient = thrust standard deviation during the forecast period.

[0068] Pitch smoothing coefficient = sum of squares of pitch angle changes.

[0069] Quadratic programming solver:

[0070] Calling industrial optimization libraries (such as CPLEX) to solve the pitch angle sequence that satisfies the minimum cost function within 0.1 seconds,

[0071] If the limit is not exceeded: keep the current angle unchanged to avoid invalid movements.

[0072] Furthermore, the second target instruction calculated based on the first target instruction and combined with the amplitude constraint and rate constraint of the blade includes screening instructions with pitch amplitude within the amplitude constraint interval and pitch speed within the rate constraint interval as the second target instruction.

[0073] For example, the amplitude constraint range is: the pitch angle is limited to [0°, 90°] (mechanical limit value),

[0074] Rate constraint range: pitch speed does not exceed 8° / s (pitch bearing limit),

[0075] Screening method: For each angle step β of the first target instruction t :

[0076] Check | Beta t -β t-1 |≤8° (single-step amplitude constraint),

[0077] Calculate |(β t -β t-1 ) / Δt|≤8° / s (rate constraint),

[0078] Eliminate out-of-limit points and generate a smooth second target instruction through linear interpolation.

[0079] Furthermore, the second operating data includes historical wind speed data, pitch angle execution sequence, actual thrust measurement value and tower vibration spectrum.

[0080] For example, the time range is the same length as the prediction period (e.g., if you want to predict the next 30 seconds, use the data from the past 30 seconds).

[0081] Data type: Historical wind speed: 1Hz sampling data recorded by SCADA,

[0082] Pitch angle sequence: pitch system execution log,

[0083] Actual thrust: wheel hub sensor historical average (after filtering),

[0084] Vibration Spectrum: A snapshot of the spectrum stored by the FFT analysis library.

[0085] Furthermore, the training process uses a long short-term memory network to model the relationship between thrust and vibration, the input layer includes wind speed, pitch angle, and tip speed ratio, and the output layer includes thrust prediction values ​​and tower vibration frequency amplitude.

[0086] Input layer processing: wind speed / pitch angle / tip speed ratio form a three-dimensional vector and form a time series queue with a step size of 0.1 seconds.

[0087] Output layer settings, thrust prediction value: thrust peak value at 3 seconds, 6 seconds, and 9 seconds in the future

[0088] Vibration amplitude: vibration acceleration amplitude of the tower frequency (0.1-0.3Hz)

[0089] Training method: Use the past 30 days of data to train the initial model.

[0090] 500 time series segments (30 seconds in length) are input in each batch and iterated 1000 times.

[0091] Furthermore, the method further includes adding vibration suppression after calculating the target instruction, and when the frequency amplitude in the tower vibration spectrum exceeds a preset threshold, superimposing an anti-phase compensation angle in the pitch angle sequence.

[0092] Trigger condition: The tower frequency amplitude is detected to be greater than the preset threshold (such as 20mm / s 2 )

[0093] Compensation angle generation: Determine the current vibration peak position through the phase analyzer,

[0094] Calculate the reverse phase angle: Δβ=-K×(AA max ),

[0095] K: Compensation gain (empirical value 0.5° / mm / s2 ),

[0096] A: Real-time vibration amplitude.

[0097] Superimpose Δβ to the corresponding time point of the target pitch angle sequence.

[0098] Furthermore, before controlling the wind turbine generator set to change the pitch according to the target instruction, a safety verification is performed to compare the difference between the pitch angle output by the prediction model and the current actual pitch angle.

[0099] Calculate the target pitch angle change Δβ=|β tar -β krt |, small angle mode (Δβ≤10°):

[0100] Send commands directly to the pitch control system, large angle mode (Δβ>10°):

[0101] Phase 1: Implementation to Beta krt +0.5×Δβ(speed limit 5° / s),

[0102] Pause for 3 seconds to monitor the thrust change rate,

[0103] If the rate of change is ≤10% of rated thrust / second, execute the remaining angle.

[0104] If the limit is exceeded, a safety shutdown is triggered.

[0105] The electronic device 200 may be a desktop computer, a notebook, a PDA, a cloud server, or other electronic device. The electronic device 200 may include but is not limited to a processor 201 and a memory 202. Those skilled in the art will appreciate that Figure 2 This is merely an example of the electronic device 200 and does not constitute a limitation of the electronic device 200. The electronic device 200 may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, the electronic device may also include input and output devices, network access devices, buses, etc.

[0106] The processor 201 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor.

[0107] The memory 202 can be an internal storage unit of the electronic device 200, such as a hard disk or memory of the electronic device 200. The memory 202 can also be an external storage device of the electronic device 200, such as a plug-in hard disk equipped on the electronic device 200, a smart memory card (SMC), a secure digital (SD) card, a flash card, etc. Furthermore, the memory 202 can also include both an internal storage unit of the electronic device 200 and an external storage device. The memory 202 is used to store the computer program 203 and other programs and data required by the electronic device. The memory 202 can also be used to temporarily store data that has been output or is about to be output.

[0108] The above embodiments are only used to illustrate the technical solutions of the present disclosure, rather than to limit them. Although the present disclosure has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present disclosure, and should all be included in the scope of protection of the present disclosure.

Claims

1. A pitch control method for a wind turbine generator system, characterized in that: The following steps are involved: Acquire first operating data of the wind turbine generator set in a first time period, where the first operating data includes wind speed time series data, hub axial thrust, pitch angle, and tower vibration spectrum of the wind turbine generator set in the first time period, where the end time of the first time period is real time; The first operation data is input into a pre-trained prediction model to calculate a target instruction, wherein the training process of the prediction model includes: Acquire second operating data of a second time period of the wind turbine generator set, where the end time of the second time period is the start time of the first time period, predicting the thrust peak value of the first time period according to the second operating data, and calculating a first target instruction based on the thrust peak value, wherein the first target instruction includes a pitch angle sequence, Based on the first target instruction, a second target instruction is calculated in combination with the amplitude constraint and the rate constraint of the blade, the aerodynamic load of the blade under the second target instruction is calculated, and the training is terminated when the aerodynamic load meets a set threshold; The wind turbine generator set is controlled to change pitch according to the target instruction.

2. The method according to claim 1, characterized in that The thrust peak value of the first time period is predicted by the following formula: Where ρ is the air density, R is the radius of the wind wheel, C p is the wind energy utilization coefficient, λ is the tip speed ratio, β is the pitch angle, V w is the wind speed.

3. The method according to claim 2, characterized in that The calculating of the first target instruction based on the thrust peak value includes calculating the optimal pitch angle when the predicted thrust peak value exceeds the design safety threshold, and otherwise maintaining the current pitch angle, specifically including: Establish a cost function with thrust fluctuation coefficient and pitch smoothness coefficient as optimization objectives; The minimum value of the cost function is solved by quadratic programming to obtain the optimal pitch angle.

4. The method according to claim 3, characterized in that The calculating the second target instruction based on the first target instruction and in combination with the amplitude constraint and the rate constraint of the blade includes screening instructions with pitch amplitudes within the amplitude constraint interval and pitch speeds within the rate constraint interval as the second target instruction.

5. The method according to claim 1, wherein The second operating data includes historical wind speed data, pitch angle execution sequence, actual thrust measurement value and tower vibration spectrum.

6. The method according to claim 1, characterized in that The training process uses a long short-term memory network to model the relationship between thrust and vibration, the input layer includes wind speed, pitch angle, and tip speed ratio, and the output layer includes thrust prediction values ​​and tower vibration frequency amplitude.

7. The method according to claim 1, characterized in that The method further includes adding vibration suppression after calculating the target command, and superimposing an anti-phase compensation angle in the pitch angle sequence when the frequency amplitude in the tower vibration spectrum exceeds a preset threshold.

8. The method according to claim 1, characterized in that Before controlling the wind turbine generator set to change the pitch according to the target instruction, a safety verification is performed to compare the difference between the pitch angle output by the prediction model and the current actual pitch angle.

9. An electronic device, characterized in that: include: one or more processors; A storage unit is used to store one or more programs, which, when executed by the one or more processors, enable the one or more processors to implement the pitch control method of the wind turbine according to any one of claims 1 to 8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the pitch control method for a wind turbine generator set according to any one of claims 1 to 8 can be implemented.

Images