Wind turbine generator vibration protection method and device and medium

Abstract

The invention discloses a wind turbine generator vibration protection method and device and a medium. Relates to the field of wind power generation and solves the safety protection problem of a wind turbine generator under the vibration condition. According to the method, acceleration estimation is carried out through the two multiple linear regression models established through the front-back acceleration signals and the left-right acceleration signals, the vibration state of the wind turbine generator is predicted in real time, the vibration prediction state is determined according to the two estimated acceleration signals, and the vibration prediction efficiency is improved under the condition that hardware cost does not need to be additionally increased. By collecting semaphores of front, back, left and right accelerations of a cabin and accurately estimating the semaphores, the wind turbine generator is protected when the wind turbine generator vibrates, so that safe and stable operation of the wind turbine generator is realized, and the service life of components of the wind turbine generator is prolonged.

Description

Technical Field

[0001] The present application relates to the field of wind power generation, and in particular to a method, device and medium for protecting a wind turbine from vibration. Background Art

[0002] As the global demand for renewable energy continues to grow, wind power has developed rapidly as a clean and efficient form of energy. However, in the process of wind power generation, due to the complexity and variability of the natural environment, wind turbines often face various challenges, among which vibration and impact are particularly prominent. Vibration not only affects the stability and life of wind turbines, but may also cause structural damage or failure, thereby affecting power generation efficiency and even causing safety accidents.

[0003] Currently, real-time vibration data is usually collected and processed through sensors installed on the cabin, but it has a lag. When an abnormality is detected, emergency processing may have already caused structural damage or failure.

[0004] It can be seen that how to solve the problem of safety protection of wind turbines under vibration conditions is a technical problem that needs to be solved urgently by people in this field. Summary of the invention

[0005] The purpose of the present application is to provide a wind turbine vibration protection method, device and medium to solve the problem of wind turbine safety protection under vibration conditions.

[0006] In order to solve the above technical problems, the present application provides a wind turbine vibration protection method, comprising:

[0007] Continuously collect the front and rear acceleration signals and the left and right acceleration signals of the cabin;

[0008] The front-rear acceleration signal and the left-right acceleration signal are obtained as independent variables, and the front-rear acceleration estimated value and the left-right acceleration estimated value are obtained as dependent variables to train a front-rear acceleration multiple linear regression model and a left-right acceleration multiple linear regression model, and obtain estimated regression coefficients;

[0009] Obtaining front and rear acceleration estimated values ​​and left and right acceleration estimated values ​​respectively according to the estimated regression coefficient, the real-time front and rear acceleration signal, and the left and right acceleration signal;

[0010] Determining whether the front-rear acceleration estimated value and the left-right acceleration estimated value meet a shutdown condition;

[0011] If satisfied, the shutdown operation is performed.

[0012] As an optional implementation scheme, in the above-mentioned wind turbine vibration protection method, the step of determining whether the front and rear acceleration estimated values ​​and the left and right acceleration estimated values ​​meet the shutdown condition includes:

[0013] Obtaining fusion data according to the front and rear acceleration estimated values ​​and the left and right acceleration estimated values;

[0014] Determine whether the front and rear acceleration estimated value is greater than a first threshold, or the left and right acceleration estimated value is greater than a second threshold, or the fusion data is greater than a third threshold;

[0015] If so, it is determined that the shutdown condition is met;

[0016] If not, it is determined that the shutdown condition is not met.

[0017] As an optional implementation scheme, in the above-mentioned wind turbine vibration protection method, before judging whether the front and rear acceleration estimated values ​​and the left and right acceleration estimated values ​​meet the shutdown condition, it also includes:

[0018] Get the preset filter value;

[0019] The sequence of the front and rear acceleration estimated values ​​and the sequence of the left and right acceleration estimated values ​​are filtered according to the preset filtering values.

[0020] As an optional implementation scheme, in the above-mentioned wind turbine vibration protection method, obtaining fusion data according to the front and rear acceleration estimated values ​​and the left and right acceleration estimated values ​​includes:

[0021] Obtaining fused data according to the front and rear acceleration estimated values, the left and right acceleration estimated values, and a first formula;

[0022] The first formula is: ;

[0023] Among them, A represents the estimated value of front and rear acceleration, B represents the estimated value of left and right acceleration, and M represents fused data.

[0024] As an optional implementation scheme, in the above-mentioned wind turbine vibration protection method, the execution of the shutdown operation includes:

[0025] Determine the target pitch rate and target generator torque according to the generator speed-pitch rate interpolation table;

[0026] The dispatched generator torque is the target generator torque, and the retraction rate of each blade is the target pitch rate.

[0027] As an optional implementation scheme, in the above-mentioned wind turbine vibration protection method, the front and rear acceleration signals and the left and right acceleration signals are used as independent variables, and the front and rear acceleration estimated values ​​and the left and right acceleration estimated values ​​are used as dependent variables to train the front and rear acceleration multiple linear regression model and the left and right acceleration multiple linear regression model, and obtain the estimated regression coefficients, including:

[0028] Every five consecutive front and back acceleration signals and the front and back acceleration signals after a preset time are used as front and back acceleration estimation values ​​to generate a set of training front and back acceleration data;

[0029] Every five consecutive left and right acceleration signals and the left and right acceleration signals after a preset time are used as left and right acceleration estimated values ​​to generate a set of training left and right acceleration data;

[0030] According to the multiple groups of acceleration data before and after the training, and the multiple groups of acceleration data about the left and right sides of the training, respectively, estimated regression coefficients are obtained;

[0031] Determine whether the model fit meets a preset threshold value according to the estimated regression coefficients;

[0032] If so, the front-to-back acceleration multiple linear regression model and the left-to-right acceleration multiple linear regression model are output.

[0033] As an optional implementation scheme, in the above-mentioned wind turbine vibration protection method, whether the model fit meets the preset threshold value will be determined according to the estimated regression coefficient, including:

[0034] Determine the mean square error and the determination coefficient of the front-rear acceleration multiple linear regression model and the left-right acceleration multiple linear regression model respectively according to the estimated regression coefficients;

[0035] Determine whether the mean square error is less than a preset error and whether the determination coefficient is greater than a preset coefficient;

[0036] If so, it is determined that the model fit meets the preset threshold.

[0037] In order to solve the above technical problems, the present application also provides a wind turbine vibration protection device, comprising:

[0038] The acquisition module is used to continuously acquire the front and rear acceleration signals and the left and right acceleration signals of the cabin;

[0039] A training module, used to obtain the front-to-back acceleration signal and the left-to-right acceleration signal as independent variables and the front-to-back acceleration estimated value and the left-to-right acceleration estimated value as dependent variables to train a front-to-back acceleration multiple linear regression model and a left-to-right acceleration multiple linear regression model, and obtain estimated regression coefficients;

[0040] An estimation module, used to obtain a front and rear acceleration estimation value and a left and right acceleration estimation value respectively according to the estimated regression coefficient, the real-time front and rear acceleration signal, and the left and right acceleration signal;

[0041] A judgment module, used to judge whether the front and rear acceleration estimated values ​​and the left and right acceleration estimated values ​​meet the shutdown condition; if so, triggering the protection module;

[0042] The protection module is used to execute a shutdown operation.

[0043] In order to solve the above technical problems, the present application also provides a wind turbine vibration protection device, comprising:

[0044] Memory for storing computer programs;

[0045] The processor is used to implement the steps of the above-mentioned wind turbine vibration protection method when executing the computer program.

[0046] In order to solve the above technical problems, the present application also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the above wind turbine vibration protection method are implemented.

[0047] The wind turbine vibration protection method provided by the present application continuously collects the front and rear acceleration signals and the left and right acceleration signals of the cabin; obtains the front and rear acceleration signals and the left and right acceleration signals as independent variables, and the front and rear acceleration estimated values ​​and the left and right acceleration estimated values ​​as dependent variables to train the front and rear acceleration multiple linear regression model and the left and right acceleration multiple linear regression model, and obtains the estimated regression coefficient; obtains the front and rear acceleration estimated values ​​and the left and right acceleration estimated values ​​according to the estimated regression coefficient, the real-time front and rear acceleration signals and the left and right acceleration signals respectively; judges whether the front and rear acceleration estimated values ​​and the left and right acceleration estimated values ​​meet the shutdown conditions; if they meet, executes the shutdown operation. The present application respectively uses two multiple linear regression models established by the front and rear acceleration signals and the left and right acceleration signals to perform acceleration estimation, predict the vibration state of the wind turbine in real time, and determine the vibration prediction state according to the two estimated acceleration signals. Without the need to increase the hardware cost, the signal quantity of the front and rear acceleration and the left and right acceleration of the cabin can be collected, and accurate estimation can be performed to protect the wind turbine when vibration occurs, thereby achieving safe and stable operation of the unit and extending the life of the unit components.

[0048] In addition, the present application also provides a device and a medium, which correspond to the above-mentioned wind turbine vibration protection method and have the same effect as above. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the embodiments of the present application, the following is a brief introduction to the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0050] Figure 1 A flow chart of a wind turbine vibration protection method provided in an embodiment of the present application;

[0051] Figure 2 A schematic diagram of a wind turbine vibration protection device provided in an embodiment of the present application;

[0052] Figure 3 A schematic diagram of another wind turbine vibration protection device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0053] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0054] The core of this application is to provide a wind turbine vibration protection method, device and medium.

[0055] In order to enable those skilled in the art to better understand the present application, the present application is further described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0056] In order to solve the above problems, the present invention provides a wind turbine vibration protection method. Figure 1 A flow chart of a wind turbine vibration protection method provided in an embodiment of the present application is shown in FIG. Figure 1 As shown, including:

[0057] S11: continuously collect the front and rear acceleration signals and the left and right acceleration signals of the cabin;

[0058] S12: Obtaining the front and rear acceleration signals and the left and right acceleration signals as independent variables and the front and rear acceleration estimated values ​​and the left and right acceleration estimated values ​​as dependent variables to train the front and rear acceleration multiple linear regression models and the left and right acceleration multiple linear regression models, and obtaining estimated regression coefficients;

[0059] S13: Obtaining front and rear acceleration estimated values ​​and left and right acceleration estimated values ​​respectively according to the estimated regression coefficient, the real-time front and rear acceleration signals, and the left and right acceleration signals;

[0060] S14: Determine whether the front and rear acceleration estimated values ​​and the left and right acceleration estimated values ​​meet the shutdown condition;

[0061] S15: If satisfied, execute shutdown operation.

[0062] Step S11: Continuously collect the front and rear acceleration signals and the left and right acceleration signals of the nacelle. Continuous collection means that after the wind turbine is started, the corresponding acceleration data is obtained from the sensors arranged in the front and rear and left and right directions of the nacelle. The front and rear acceleration signals are the acceleration of vibration in the front and rear directions, and the left and right acceleration signals are the acceleration of vibration in the left and right directions.

[0063] The front and rear acceleration signals and left and right acceleration signals of the nacelle refer to the acceleration changes caused by factors such as wind force and mechanical vibration during the operation of the wind turbine. These signals can be specifically the voltage or digital signal output by the acceleration sensor. They are not limited to a specific sensor type and can be mechanical, photoelectric or micro-electro mechanical system (MEMS) sensors.

[0064] Step S12 obtains two multiple linear regression models and obtains corresponding estimated regression coefficients. The multiple linear regression model refers to using statistical methods to establish a linear model for the relationship between multiple independent variables (front and rear acceleration signals, left and right acceleration signals) and dependent variables (front and rear acceleration estimated values, left and right acceleration estimated values).

[0065] The estimated regression coefficient refers to the parameter describing the relationship between the independent variable and the dependent variable obtained by training the model with historical data, which can be specifically the slope and intercept in the model.

[0066] In one application, the training data used to train the multivariate linear regression model is data collected within a period of time after the wind turbine is started, or it can be historical data of the current wind turbine operation. The acceleration within a period of time is used as the independent variable, and the acceleration after a preset time during this period is used as the dependent variable to train the model. It should be noted that the acceleration data mentioned here are divided into two types, which are processed separately and two models are trained. (For the convenience of description, they are collectively referred to as multivariate linear regression models in the following text).

[0067] Exemplarily, every five consecutive sampling data are used as independent variables, and the acceleration after a preset time is used as a set of training data to establish a multivariate linear regression model.

[0068] Step S13 obtains the front and rear acceleration estimation values ​​and the left and right acceleration estimation values ​​according to the estimated regression coefficients. The estimation value refers to the acceleration value predicted at a certain moment in the future based on the current acceleration signal and the trained model.

[0069] The real-time front and rear acceleration signals and left and right acceleration signals refer to the acceleration data of the wind turbine at the current moment.

[0070] In this step, the future acceleration is predicted by inputting the real-time acceleration signal into the trained multivariate linear regression model.

[0071] It should be noted that in order to respond to the vibration changes of the wind turbine generator set in a timely manner, real-time prediction is performed through step S13.

[0072] Step S14: Determine whether the shutdown condition is met. The shutdown condition refers to determining whether the wind turbine needs to be shut down to avoid potential structural damage based on the estimated value and a preset safety threshold.

[0073] The front and rear acceleration estimated values ​​and the left and right acceleration estimated values ​​are compared with the predicted values ​​and the safety threshold value to determine whether the wind turbine needs to be shut down. Multiple threshold conditions and multiple judgment conditions can be set, and this embodiment does not make specific restrictions.

[0074] If not, then continue monitoring, if satisfied, in order to minimize the damage caused by vibration, step S15 performs a shutdown operation, and the shutdown operation refers to the measures taken to safely stop the operation of the wind turbine. In this step, a shutdown command is sent by the control system to achieve an emergency shutdown of the wind turbine.

[0075] The wind turbine vibration protection method provided by the embodiment of the present application continuously collects the front and rear acceleration signals and the left and right acceleration signals of the cabin; the front and rear acceleration signals and the left and right acceleration signals are obtained as independent variables, and the front and rear acceleration estimated values ​​and the left and right acceleration estimated values ​​are used as dependent variables to train the front and rear acceleration multiple linear regression model and the left and right acceleration multiple linear regression model, and obtain the estimated regression coefficient; the front and rear acceleration estimated values ​​and the left and right acceleration estimated values ​​are obtained according to the estimated regression coefficient and the real-time front and rear acceleration signals and the left and right acceleration signals respectively; it is determined whether the front and rear acceleration estimated values ​​and the left and right acceleration estimated values ​​meet the shutdown conditions; if they meet, the shutdown operation is executed. The present application uses the multiple linear regression model established by the acceleration signal to perform acceleration estimation, and predicts the vibration state of the wind turbine in real time. Without the need to increase the hardware cost, the signal quantity of the front and rear acceleration and the left and right acceleration of the cabin can be collected, and accurate estimation can be performed to protect the wind turbine when vibration occurs, thereby achieving safe and stable operation of the unit and extending the life of the unit components.

[0076] According to the above embodiment, in a specific embodiment, determining whether the front and rear acceleration estimated values ​​and the left and right acceleration estimated values ​​meet the shutdown condition includes:

[0077] The fused data is obtained according to the estimated values ​​of front and rear accelerations and the estimated values ​​of left and right accelerations;

[0078] Determine whether the front and rear acceleration estimates are greater than a first threshold, or the left and right acceleration estimates are greater than a second threshold, or the fused data is greater than a third threshold;

[0079] If so, it is determined that the shutdown condition is met;

[0080] If not, it is determined that the shutdown condition is not met.

[0081] Fusion data refers to combining the estimated front and rear accelerations and the estimated left and right accelerations through a mathematical method to form a comprehensive vibration index. This fusion can be a simple arithmetic average, weighted average, maximum value, or a more complex algorithm such as principal component analysis (PCA).

[0082] The first threshold and the second threshold are pre-set safety limits, which are used to respectively determine whether the estimated values ​​of the front and rear accelerations and the estimated values ​​of the left and right accelerations exceed a normal range.

[0083] The third threshold is a safety limit set for the fused data, and is used to determine whether a shutdown is required by comprehensively considering the front and rear accelerations and the left and right accelerations. If any one of the conditions is met: the front and rear acceleration estimate is greater than the first threshold, or the left and right acceleration estimate is greater than the second threshold, or the fused data is greater than the third threshold, then the shutdown condition is determined to be met.

[0084] If the shutdown conditions are met, the shutdown operation is performed to protect the wind turbine from possible damage. If the shutdown conditions are not met, the wind turbine continues to operate normally.

[0085] By comprehensively considering the estimated acceleration values ​​in multiple directions, the vibration state of the wind turbine can be more comprehensively evaluated, thereby improving the safety of shutdown decisions. Fusion data can reduce the misjudgment that may be caused by the estimated acceleration value in a single direction and improve the accuracy of shutdown decisions. Accurate shutdown decisions can reduce unnecessary shutdowns, thereby reducing maintenance costs and improving the availability of wind turbines.

[0086] According to the above embodiment, in a specific embodiment, the wind turbine vibration protection method, before determining whether the front and rear acceleration estimated values ​​and the left and right acceleration estimated values ​​meet the shutdown condition, further includes:

[0087] Get the preset filter value;

[0088] The sequence of front and rear acceleration estimation values ​​and the sequence of left and right acceleration estimation values ​​are filtered according to the preset filtering values.

[0089] The sequences of the front and rear acceleration estimation values ​​and the sequences of the left and right acceleration estimation values ​​are filtered respectively, and the filtering parameters are preset according to the operating characteristics and environmental conditions of the wind turbine generator set.

[0090] The sequence of front and rear acceleration estimates and the sequence of left and right acceleration estimates refer to the acceleration prediction values ​​of the wind turbine in a continuous time, forming time series data. The acceleration estimate sequence is filtered according to the preset filter value to reduce the influence of measurement noise and environmental interference and improve the accuracy of the estimate.

[0091] Through filtering, high-frequency noise and interference in the acceleration signal can be removed, the signal quality can be improved, and more accurate data can be provided for subsequent prediction and judgment. The estimated value sequence after filtering is smoother and more stable, which helps to improve the prediction accuracy of the multivariate linear regression model. Accurate estimated values ​​can reduce misjudgments caused by noise, reduce unnecessary downtime risks, and improve the operating efficiency of wind turbines.

[0092] According to the above embodiment, in a specific embodiment, the fusion data is obtained according to the front and rear acceleration estimation values ​​and the left and right acceleration estimation values, including:

[0093] The fused data is obtained according to the front and rear acceleration estimated values, the left and right acceleration estimated values, and the first formula;

[0094] The first formula is: ;

[0095] Among them, A represents the estimated value of front and rear acceleration, B represents the estimated value of left and right acceleration, and M represents fused data.

[0096] In this embodiment, the square of the front and rear acceleration estimates (A) and the square of the left and right acceleration estimates (B) are added, and the square root of the sum is taken to obtain fused data. The fused data (M) provides a measure of the comprehensive vibration intensity of the wind turbine in two vertical directions, which helps to more comprehensively evaluate the vibration state of the turbine.

[0097] By combining vibration information from two directions, the deviation that may be caused by vibration data from a single direction can be reduced, and the accuracy of shutdown decisions can be improved. The use of fused data simplifies the judgment logic of shutdown conditions, and operators can more easily determine whether a shutdown operation is required based on a single indicator.

[0098] According to the above embodiment, in a specific embodiment, performing a shutdown operation includes:

[0099] Determine the target pitch rate and target generator torque according to the generator speed-pitch rate interpolation table;

[0100] The dispatched generator torque is the target generator torque, and the retraction rate of each blade is the target pitch rate.

[0101] The generator speed-pitch rate interpolation table is a preset table that determines the target pitch rate and target generator torque that should be achieved during the shutdown process based on the current working state of the wind turbine (such as the current generator speed).

[0102] The target pitch rate refers to the rate of change of the blade angle that each blade needs to be adjusted to in order to achieve a safe shutdown. The target generator torque refers to the torque value that the generator should maintain during the shutdown process to control the speed drop.

[0103] Issue control instructions to adjust the actual operating parameters of the wind turbine to match the target parameters. The generator torque can be adjusted by changing the excitation current of the generator or other control methods. The adjustment of the retraction rate of each blade is achieved through the pitch control system to ensure that the blade moves to the predetermined position according to the target pitch rate.

[0104] By precisely controlling the generator torque and pitch rate, the wind turbine can be shut down safely and smoothly, avoiding damage caused by sudden shutdown. Smooth shutdown operation reduces wear and tear on mechanical parts and extends the service life of the wind turbine.

[0105] Operators can monitor the shutdown process in real time through the remote monitoring system and intervene manually when necessary. After shutdown, a series of inspection and maintenance procedures can be automatically triggered to ensure that the wind turbine is in the best condition before the next startup.

[0106] According to the above embodiment, in a specific embodiment, the front and rear acceleration signals and the left and right acceleration signals are used as independent variables, and the front and rear acceleration estimated values ​​and the left and right acceleration estimated values ​​are used as dependent variables to train the front and rear acceleration multiple linear regression model and the left and right acceleration multiple linear regression model, and obtain the estimated regression coefficients, including:

[0107] Every five consecutive front and back acceleration signals and the front and back acceleration signals after a preset time are used as front and back acceleration estimation values ​​to generate a set of training front and back acceleration data;

[0108] Every five consecutive left and right acceleration signals and the left and right acceleration signals after a preset time are used as left and right acceleration estimated values ​​to generate a set of training left and right acceleration data;

[0109] The estimated regression coefficients were obtained based on multiple groups of acceleration data before and after training, and multiple groups of acceleration data between left and right training;

[0110] Determine whether the model fit meets the preset threshold value according to the estimated regression coefficient;

[0111] If so, the multiple linear regression model of front and rear acceleration and the multiple linear regression model of left and right acceleration are output.

[0112] Taking the front and back acceleration as an example, five consecutive data points are selected from the collected front and back acceleration signals. After the above five data points, an acceleration signal at a preset time point is selected as the estimated value of the front and back acceleration. These six data points (five consecutive signals and one estimated value) are combined into a set of training data: ,in, . i represents the i-th group of training data; Represents the 5 front and rear acceleration signal data of the i-th group of data; Represents the estimated acceleration before and after the i-th group of data.

[0113] Establish the before and after multiple linear regression model: ;

[0114] Similarly, a left-right multivariate linear regression model is established based on multiple groups of training left-right acceleration data: .

[0115] in, and represents the intercept term, , represents the regression coefficient, is the error term. represents the independent variable, a set of front and rear acceleration signals; Represents the independent variable, a set of left and right acceleration signals.

[0116] The error term represents the deviation between the actual value and the value predicted by the regression model, and It follows a normal distribution with mean 0.

[0117] The estimated regression coefficient is calculated using the least squares method to minimize the sum of squared errors. Since the front and rear accelerations and left and right accelerations of the cabin are processed in the same way, only the front and rear accelerations of the cabin are used as an example, that is:

[0118] ;

[0119] in, represents a set of regression coefficients, i represents the i-th group of training data; Represents the 5 front and rear acceleration signal data of the i-th group of data; Represents the estimated acceleration before and after the i-th group of data.

[0120] In order to find the minimum Q , take partial derivatives of them respectively, and set the partial derivatives equal to 0, we can get a normal system of equations, which can be solved by using matrix form.

[0121] make , , ;

[0122] Then the normal equations can be expressed as The estimated value of the regression coefficient is

[0123] .

[0124] After obtaining the estimated value of the regression coefficient, the fitted value of the dependent variable A can be calculated :

[0125] ;

[0126] After determining the estimated regression coefficient, the fitting value is calculated through the preset test data, and the error index is further calculated. The training effect of the current regression model is judged based on the fitting value and the actual estimated value. Cross-validation technology can be used to evaluate the generalization ability and stability of the model to ensure that the model performs consistently on different data sets.

[0127] Specifically, whether the model fit meets the preset threshold will be determined based on the estimated regression coefficients, including:

[0128] According to the estimated regression coefficients, the mean square error and determination coefficient of the multiple linear regression model of front and rear acceleration and the multiple linear regression model of left and right acceleration were determined respectively;

[0129] Determine whether the mean square error is less than the preset error and whether the determination coefficient is greater than the preset coefficient;

[0130] If so, it is determined that the model fit meets the preset threshold.

[0131] Determine the mean square error by calculating the formula ; and the coefficient of determination .

[0132] The smaller the mean square error, the larger the determination coefficient, and the better the training effect of the model. The specific preset error and preset coefficient are determined according to actual needs.

[0133] By evaluating the mean square error and the coefficient of determination, the prediction performance of the model can be quantified and the reliability and credibility of the model can be improved. Based on the evaluation results, the model parameters can be adjusted, the model structure can be optimized, and the prediction accuracy of the model can be improved. An accurate model can reduce unnecessary downtime and maintenance, reduce maintenance costs, and improve the availability of wind turbines.

[0134] In the above embodiments, the wind turbine vibration protection method is described in detail, and the present application also provides corresponding embodiments of the wind turbine vibration protection device. It should be noted that the present application describes the embodiments of the device part from two perspectives, one is based on the functional module perspective, and the other is based on the hardware perspective.

[0135] From the perspective of functional modules, Figure 2 A structural diagram of a wind turbine vibration protection device provided in an embodiment of the present application, such as Figure 2 As shown, a wind turbine vibration protection device comprises:

[0136] The acquisition module 21 is used to continuously acquire the front and rear acceleration signals and the left and right acceleration signals of the cabin;

[0137] The training module 22 is used to obtain the front and rear acceleration signals and the left and right acceleration signals as independent variables, and the front and rear acceleration estimated values ​​and the left and right acceleration estimated values ​​as dependent variables to train the front and rear acceleration multiple linear regression model and the left and right acceleration multiple linear regression model, and obtain the estimated regression coefficients;

[0138] An estimation module 23, used to obtain a front and rear acceleration estimation value and a left and right acceleration estimation value respectively according to an estimated regression coefficient, a real-time front and rear acceleration signal, and a left and right acceleration signal;

[0139] The judging module 24 is used to judge whether the estimated values ​​of the front and rear accelerations and the estimated values ​​of the left and right accelerations meet the shutdown conditions; if they do, the protection module 25 is triggered;

[0140] The protection module 25 is used to execute the shutdown operation.

[0141] Since the embodiments of the apparatus part correspond to the embodiments of the method part, please refer to the description of the embodiments of the method part for the embodiments of the apparatus part, which will not be repeated here.

[0142] Figure 3 A structural diagram of another wind turbine vibration protection device provided in an embodiment of the present application, such as Figure 3 As shown, the wind turbine vibration protection device comprises: a memory 30 for storing a computer program;

[0143] The processor 31 is used to implement the steps of the method for obtaining user operation habit information in the above embodiment (wind turbine vibration protection method) when executing the computer program.

[0144] The wind turbine vibration protection device provided in this embodiment may include but is not limited to a smart phone, a tablet computer, a laptop computer or a desktop computer.

[0145] Among them, the processor 31 may include one or more processing cores, such as a 4-core processor, an 8-core processor, etc. The processor 31 may be implemented in at least one hardware form of a digital signal processor (DSP), a field-programmable gate array (FPGA), and a programmable logic array (PLA). The processor 31 may also include a main processor and a coprocessor. The main processor is a processor for processing data in the awake state, also known as a central processing unit (CPU); the coprocessor is a low-power processor for processing data in the standby state. In some embodiments, the processor 31 may be integrated with a graphics processing unit (GPU), and the GPU is responsible for rendering and drawing the content to be displayed on the display screen. In some embodiments, the processor 31 may also include an artificial intelligence (AI) processor, which is used to process computing operations related to machine learning.

[0146] The memory 30 may include one or more computer-readable storage media, which may be non-transitory. The memory 30 may also include a high-speed random access memory, and a non-volatile memory, such as one or more disk storage devices, flash memory storage devices. In this embodiment, the memory 30 is at least used to store the following computer program 301, wherein, after the computer program is loaded and executed by the processor 31, it can implement the relevant steps of the wind turbine vibration protection method disclosed in any of the aforementioned embodiments. In addition, the resources stored in the memory 30 may also include an operating system 302 and data 303, etc., and the storage method may be temporary storage or permanent storage. Among them, the operating system 302 may include Windows, Unix, Linux, etc. Data 303 may include but is not limited to data involved in implementing the wind turbine vibration protection method, etc.

[0147] In some embodiments, the wind turbine vibration protection device may further include a display screen 32 , an input / output interface 33 , a communication interface 34 , a power supply 35 , and a communication bus 36 .

[0148] Those skilled in the art will understand that Figure 3 The structure shown in the figure does not constitute a limitation on the wind turbine vibration protection device, and may include more or fewer components than those shown in the figure.

[0149] The wind turbine vibration protection device provided in the embodiment of the present application includes a memory and a processor. When the processor executes the program stored in the memory, it can implement the following method: wind turbine vibration protection method.

[0150] Finally, the present application also provides an embodiment corresponding to a computer-readable storage medium. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps described in the above-mentioned embodiment of the wind turbine vibration protection method are implemented.

[0151] It is understandable that if the method in the above embodiment is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk and other media that can store program code.

[0152] The computer-readable storage medium provided in this embodiment stores a computer program, and when a processor executes the program, the following method can be implemented: a wind turbine vibration protection method.

[0153] The wind turbine vibration protection method, device and medium provided by the present application are introduced in detail above. The various embodiments in the specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same and similar parts between the embodiments can be referenced to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description. It should be pointed out that for ordinary technicians in this technical field, without departing from the principles of the present application, several improvements and modifications can be made to the present application, and these improvements and modifications also fall within the scope of protection of the claims of the present application.

[0154] It should also be noted that, in this specification, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "comprise", "include" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the statement "comprises a ..." does not exclude the presence of other identical elements in the process, method, article or device including the element.

Claims

1. A wind turbine vibration protection method, characterized in that: include: Continuously collect the front and rear acceleration signals and the left and right acceleration signals of the cabin; The front-rear acceleration signal and the left-right acceleration signal are obtained as independent variables, and the front-rear acceleration estimated value and the left-right acceleration estimated value are obtained as dependent variables to train a front-rear acceleration multiple linear regression model and a left-right acceleration multiple linear regression model, and obtain estimated regression coefficients; Obtaining front and rear acceleration estimated values ​​and left and right acceleration estimated values ​​respectively according to the estimated regression coefficient, the real-time front and rear acceleration signal, and the left and right acceleration signal; Determining whether the front-rear acceleration estimated value and the left-right acceleration estimated value meet a shutdown condition; If satisfied, the shutdown operation is performed.

2. The wind turbine vibration protection method according to claim 1, characterized in that: The determining whether the front and rear acceleration estimated values ​​and the left and right acceleration estimated values ​​meet the shutdown condition includes: Obtaining fusion data according to the front and rear acceleration estimated values ​​and the left and right acceleration estimated values; Determine whether the front and rear acceleration estimated value is greater than a first threshold, or the left and right acceleration estimated value is greater than a second threshold, or the fusion data is greater than a third threshold; If so, it is determined that the shutdown condition is met; If not, it is determined that the shutdown condition is not met.

3. The wind turbine vibration protection method according to claim 1, characterized in that: Before determining whether the front and rear acceleration estimated values ​​and the left and right acceleration estimated values ​​meet the shutdown condition, the method further includes: Get the preset filter value; The sequence of the front and rear acceleration estimated values ​​and the sequence of the left and right acceleration estimated values ​​are filtered according to the preset filtering values.

4. The wind turbine vibration protection method according to claim 2, characterized in that: The obtaining of fusion data according to the front and rear acceleration estimates and the left and right acceleration estimates includes: Obtaining fused data according to the front and rear acceleration estimated values, the left and right acceleration estimated values, and a first formula; The first formula is: ; Among them, A represents the estimated value of front and rear acceleration, B represents the estimated value of left and right acceleration, and M represents fused data.

5. The wind turbine vibration protection method according to claim 1, characterized in that: The execution of the shutdown operation includes: Determine the target pitch rate and target generator torque according to the generator speed-pitch rate interpolation table; The dispatched generator torque is the target generator torque, and the retraction rate of each blade is the target pitch rate.

6. The wind turbine vibration protection method according to claim 1, characterized in that: The front and rear acceleration signals and the left and right acceleration signals are used as independent variables, and the front and rear acceleration estimated values ​​and the left and right acceleration estimated values ​​are used as dependent variables to train the front and rear acceleration multiple linear regression model and the left and right acceleration multiple linear regression model, and obtain the estimated regression coefficients, including: Every five consecutive front and back acceleration signals and the front and back acceleration signals after a preset time are used as front and back acceleration estimation values ​​to generate a set of training front and back acceleration data; Every five consecutive left and right acceleration signals and the left and right acceleration signals after a preset time are used as left and right acceleration estimated values ​​to generate a set of training left and right acceleration data; According to the multiple groups of acceleration data before and after the training, and the multiple groups of acceleration data about the left and right sides of the training, respectively, estimated regression coefficients are obtained; Determine whether the model fit meets a preset threshold value according to the estimated regression coefficients; If so, the front-to-back acceleration multiple linear regression model and the left-to-right acceleration multiple linear regression model are output.

7. The wind turbine vibration protection method according to claim 6, characterized in that: Whether the model fit meets the preset threshold is determined according to the estimated regression coefficients, including: Determine the mean square error and the determination coefficient of the front-rear acceleration multiple linear regression model and the left-right acceleration multiple linear regression model respectively according to the estimated regression coefficients; Determine whether the mean square error is less than a preset error and whether the determination coefficient is greater than a preset coefficient; If so, it is determined that the model fit meets the preset threshold.

8. A wind turbine vibration protection device, characterized in that: include: The acquisition module is used to continuously acquire the front and rear acceleration signals and the left and right acceleration signals of the cabin; A training module, used to obtain the front-to-back acceleration signal and the left-to-right acceleration signal as independent variables and the front-to-back acceleration estimated value and the left-to-right acceleration estimated value as dependent variables to train a front-to-back acceleration multiple linear regression model and a left-to-right acceleration multiple linear regression model, and obtain estimated regression coefficients; An estimation module, used to obtain a front and rear acceleration estimation value and a left and right acceleration estimation value respectively according to the estimated regression coefficient, the real-time front and rear acceleration signal, and the left and right acceleration signal; A judgment module, used to judge whether the front and rear acceleration estimated values ​​and the left and right acceleration estimated values ​​meet a shutdown condition; If satisfied, the protection module is triggered; The protection module is used to execute a shutdown operation.

9. A wind turbine vibration protection device, characterized in that: include: Memory for storing computer programs; A processor is used to implement the steps of the wind turbine vibration protection method according to any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the wind turbine vibration protection method according to any one of claims 1 to 7 are implemented.

Images