A Fault Early Warning Method for Wind Turbine Combining Load Analysis

Abstract

The present invention discloses a fault early warning method for a wind turbine combined with load analysis, which relates to the technical field of wind turbine monitoring. The method includes: collecting real-time wind force data and unit structure parameters of a target wind turbine; obtaining the wind direction change characteristics of the target wind turbine; establishing a load calculation model to obtain the load analysis result of the target wind turbine; constructing a fault early warning model; and generating a fault early warning signal. It solves the technical problem that existing wind turbine fault early warning is difficult to accurately capture and evaluate the influence of wind direction on the unit load, resulting in lagging fault early warning and limited accuracy, and insufficient reliability and efficiency of the wind turbine, and achieves the technical effects of improving the accuracy and timeliness of fault early warning and enhancing the operation stability and reliability of the unit.

Description

Technical Field

[0001] This application relates to the technical field of wind turbine monitoring, and particularly to a wind turbine fault warning method combined with load analysis. Background Art

[0002] With the continuous growth of the global demand for renewable energy, wind power generation, as a clean and renewable energy form, has become increasingly important. As the core equipment of wind power generation, the operation stability and reliability of wind turbines are directly related to the power generation efficiency and economic benefits of the entire wind farm. However, wind turbines are long-term exposed to complex and changeable natural environments and are subjected to various loads, especially the change of wind direction. Traditional wind turbine fault warning methods mainly rely on the real-time monitoring of the operation parameters of the turbines and the statistical analysis of the fault history data. Although they can identify the abnormal states of the turbines to a certain extent, there are often problems such as late warning and insufficient accuracy. Especially when facing complex and changeable wind direction changes, it is difficult to accurately capture the stress changes and potential fault risks inside the turbine structure, thus affecting the stability and lifespan of wind turbines.

[0003] In the current related technologies of wind turbine fault warning, there are technical problems that it is difficult to accurately capture and evaluate the influence of wind direction on the load of the turbine, resulting in late and limited accuracy of fault warning, and insufficient reliability and efficiency of wind turbines. Summary of the Invention

[0004] By providing a wind turbine fault warning method combined with load analysis, this application solves the technical problems existing in the current wind turbine fault warning, that is, it is difficult to accurately capture and evaluate the influence of wind direction on the load of the turbine, resulting in late and limited accuracy of fault warning, and insufficient reliability and efficiency of wind turbines, and achieves the technical effects of improving the accuracy and timeliness of fault warning and enhancing the operation stability and reliability of the turbine.

[0005] This application provides a wind turbine fault warning method combined with load analysis, including: collecting the real-time wind force data and the unit structure parameters of the target wind turbine; analyzing the influence of wind direction on the load of the target wind turbine based on the real-time wind force data to obtain the wind direction change characteristics of the target wind turbine; establishing a load calculation model according to the unit structure parameters and the wind direction change characteristics to obtain the load analysis result of the target wind turbine; constructing a fault warning model; and if there is an abnormal load change in the load analysis result, generating a fault warning signal according to the fault warning model.

[0006] In a possible implementation, when analyzing the impact of wind direction on the loads of the target wind turbine, the following processing is also performed: The wind direction data at the location of the target wind turbine is collected in real time by a wind direction sensor; The wind direction data is dynamically analyzed to obtain the wind direction change characteristics, where the wind direction change characteristics include the wind direction fluctuation amplitude, the wind direction change frequency, and the wind direction change rate; Based on the wind direction change characteristics, the load change during wind direction change is identified.

[0007] In a possible implementation, the wind turbine fault warning method combining load analysis also performs the following processing: Multiple wind speed data at different heights of the target wind turbine are collected by a wind speed sensor to obtain a wind speed gradient; The wind direction change characteristics are corrected according to the wind speed gradient.

[0008] In a possible implementation, the wind turbine fault warning method combining load analysis also performs the following processing: The ambient temperature data and ambient humidity data of the target wind turbine are collected; Based on the ambient temperature data and the ambient humidity data, an environmental impact factor of the target wind turbine is established; The load calculation model is optimized according to the environmental impact factor, and the load analysis result is adjusted.

[0009] In a possible implementation, the wind turbine fault warning method combining load analysis also performs the following processing: The real-time power output parameters of the target wind turbine are obtained; The real-time power output parameters and the load analysis result are analyzed in association to evaluate the operation efficiency of the target wind turbine; If the operation efficiency of the target wind turbine decreases, an operation and maintenance warning signal is generated.

[0010] In a possible implementation, when constructing the fault warning model, the following processing is also performed: The key components of the target wind turbine include blades, a tower, and a main shaft; The historical fault records of the target wind turbine are obtained to determine the blade fault data samples, the tower fault data samples, and the main shaft fault data samples; The first fault warning sub-model, the second fault warning sub-model, and the third fault warning sub-model are respectively constructed according to the blade fault data samples, the tower fault data samples, and the main shaft fault data samples; The fault warning model is synthesized with the first fault warning sub-model, the second fault warning sub-model, and the third fault warning sub-model.

[0011] In a possible implementation manner, a first fault warning sub-model, a second fault warning sub-model, and a third fault warning sub-model are respectively constructed according to the blade fault data sample, the tower fault data sample, and the main shaft fault data sample, and the following processing is also performed: Extract the blade load data sample and the blade fault data sample from the blade fault data sample; Use the blade load data sample as the training input and the blade fault data sample as the training output, and based on a neural network model, train to obtain the first fault warning sub-model; Similarly, train to obtain the second fault warning sub-model and the third fault warning sub-model.

[0012] In a possible implementation manner, if there is an abnormal load change in the load analysis result, a fault warning signal is generated according to the fault warning model, and the following processing is also performed: If the blade load of the target wind turbine exceeds the first load threshold, the first fault warning model generates a first fault warning signal; If the tower load of the target wind turbine exceeds the second load threshold, the second fault warning model generates a second fault warning signal; If the main shaft load of the target wind turbine exceeds the third load threshold, the third fault warning model generates a third fault warning signal.

[0013] A wind turbine fault warning method combining load analysis proposed by this application is intended to collect real-time wind data and unit structure parameters of a target wind turbine; Obtain the wind direction change characteristics of the target wind turbine; Establish a load calculation model to obtain the load analysis result of the target wind turbine; Construct a fault warning model; Generate a fault warning signal. It solves the technical problem that the existing wind turbine fault warning is difficult to accurately capture and evaluate the influence of the wind direction on the unit load, resulting in a lag in fault warning and limited accuracy, and insufficient reliability and efficiency of the wind turbine, and achieves the technical effect of improving the accuracy and timeliness of fault warning and enhancing the operation stability and reliability of the unit. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings of the embodiments of the present invention will be briefly introduced below. Flowcharts are used in this application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the operations before or below do not necessarily need to be executed precisely in sequence. On the contrary, according to needs, they can be executed in reverse order or simultaneously. At the same time, other operations can also be added to these processes, or one or several operations can be removed from these processes.

[0015] Figure 1 It is a schematic flowchart of a wind turbine fault warning method combining load analysis provided by an embodiment of the present application;

[0016] Figure 2Schematic diagram of the process for analyzing the influence of loads in a wind turbine fault warning method provided by an embodiment of the present application in combination with load analysis. Detailed implementation manners

[0017] The above description is only an overview of the technical solution of the present application. In order to be able to understand the technical means of the present application more clearly, it can be implemented according to the content of the specification. And in order to make the above and other purposes, features and advantages of the present application more obvious and understandable, the following specifically gives the detailed implementation manners of the present application.

[0018] In order to make the purpose, technical solution and advantages of the present application clearer, the present application will be further described in detail below with reference to the accompanying drawings. The described embodiments should not be regarded as limitations of the present application. All other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present application.

[0019] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments. However, it can be understood that "some embodiments" can be the same subset or different subsets of all possible embodiments, and can be combined with each other without conflict. The terms "first / second" involved are only used to distinguish similar objects and do not represent a specific order for the objects. The terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or server including a series of steps or units does not have to be limited to those steps or units clearly listed, but may include other steps or modules not clearly listed or inherent to these processes, methods, products or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the technical field to which the present application belongs. The terms used herein are only for the purpose of describing the embodiments of the present application.

[0020] An embodiment of the present application provides a wind turbine fault warning method in combination with load analysis, as Figure 1 shown. The method includes:

[0021] Step S100, collecting real-time wind power data and unit structure parameters of a target wind turbine.

[0022] Preferably, real-time wind force data and unit structure parameters of the target wind turbine are obtained. The real-time wind force data refers to the relevant data of the wind environment in which the wind turbine is located during operation, mainly including wind speed data and wind direction data. Among them, the magnitude of the wind speed directly affects the power generation efficiency and load state of the wind turbine, and the change of the wind direction will cause the change of the force direction of the wind turbine, thereby affecting its load distribution; the unit structure parameters refer to the physical characteristics and structural dimensions of the wind turbine itself, including the rotor diameter, the outer diameter when the rotor rotates, which determines the size of the rotor swept area and the length of the blades; the number of blades, the number of blades on the rotor, and different numbers of blades have different effects on the wind energy capture efficiency and load distribution; the blade length, material and shape, the blade length determines its swept area, the material selection affects the weight, strength and durability of the blade, and the shape determines the aerodynamic performance of the blade; the tower height, the tower height determines the distance of the wind turbine from the ground, thereby affecting its windward area and load state; the transmission system parameters such as the parameters of components such as the gearbox and generator determine the transmission efficiency and load transfer characteristics of the wind turbine.

[0023] Step S200, based on the analysis of the real-time wind force data, analyze the load influence of the wind direction on the target wind turbine, and obtain the wind direction change characteristics of the target wind turbine.

[0024] Preferably, the wind speed data and wind direction data of the area where the target wind turbine is located are obtained in real time through sensors installed on the wind turbine, and these data are deeply analyzed to understand the change law of the wind direction and its influence on the load of the wind turbine, including analyzing characteristics such as the frequency distribution, duration, and change speed of the wind direction and how these characteristics affect the force state of the wind turbine. Specifically, the change of the wind direction will directly affect the force state of the wind turbine. When the wind direction changes, the blades of the wind turbine will be subjected to forces and torques in different directions, resulting in changes in the overall load of the unit. On the basis of deeply analyzing the real-time wind force data and load influence, further obtain the wind direction change characteristics of the target wind turbine, including the dominant wind direction, that is, the wind direction with the highest frequency of occurrence within a certain period of time, the wind direction change range, that is, the range within which the wind direction fluctuates within a certain period of time. The larger the wind direction change range, the greater the load fluctuation of the wind turbine may be, and the wind direction change speed, that is, the speed at which the wind direction changes in a short period of time. A rapid change in the wind direction may lead to a sharp change in the force state of the wind turbine, increasing the risk of damage to the unit.

[0025] Such as Figure 2As shown, step S200 further includes step S210 of using a wind direction sensor to collect in real time the wind direction data at the location where the target wind turbine is located; step S220 of dynamically analyzing the wind direction data to obtain the wind direction change characteristics, where the wind direction change characteristics include the wind direction fluctuation amplitude, the wind direction change frequency, and the wind direction change rate; and step S230 of identifying the load change when the wind direction changes based on the wind direction change characteristics.

[0026] Preferably, a wind direction sensor is a device for measuring the wind direction. It can capture and convert the wind direction information into electrical signals or digital data in real time. In the application scenario of wind turbines, the wind direction sensor is usually installed at an appropriate position of the wind turbine to collect in real time the wind direction data at the location of the unit. The collected wind direction data needs to be dynamically analyzed to extract the wind direction change characteristics, including the wind direction fluctuation amplitude, the wind direction change frequency, and the wind direction change rate. Specifically, the wind direction fluctuation amplitude refers to the maximum degree to which the wind direction deviates from the average wind direction within a certain period of time, reflecting the stability or volatility of the wind direction. The wind direction change frequency refers to the number of times the wind direction changes within a certain period of time, reflecting the speed of the wind direction change. The wind direction change rate refers to the amount of change in the wind direction per unit time, reflecting the speed and trend of the wind direction change. By dynamically analyzing the wind direction data, these wind direction change characteristics can be obtained, providing a basis for subsequent identification of load changes; then, based on the wind direction change characteristics, the load change when the wind direction changes is identified. Specifically, the load change of the wind turbine is closely related to the wind direction change. When the wind direction changes, key components such as the blades and tower of the wind turbine will be affected by different wind forces and wind pressures, resulting in load changes. Based on the wind direction change characteristics, the load change when the wind direction changes can be further identified, specifically including establishing a mathematical model between the wind direction change and the load change according to the structural characteristics and historical data of the wind turbine. While real-time monitoring the wind direction data, the load data of the wind turbine is also monitored in real time. The wind direction change characteristics are compared and analyzed with the load data to identify the load change law when the wind direction changes.

[0027] Step S220 further includes step S221 of collecting a plurality of wind speed data at different heights of the target wind turbine through a wind speed sensor to obtain a wind speed gradient; and step S222 of correcting the wind direction change characteristics according to the wind speed gradient.

[0028] Preferably, wind speed sensors (such as gradient anemometers) are installed at different heights of the target wind turbine, and multiple wind speed data at each height are collected in real time. Specifically, the wind speed sensors are usually installed on the tower of the wind turbine or in the surrounding environment to measure the wind speed at different heights. By collecting wind speed data at multiple different heights, a gradient curve of wind speed with height (i.e., wind speed gradient) can be constructed, which reflects the distribution and change trend of wind speed at different heights of the target wind turbine. During the actual operation process, the wind direction change characteristics may be affected by various factors, including the influence of the wind speed gradient. Therefore, the wind direction change characteristics are corrected according to the wind speed gradient to more accurately describe and predict the wind direction change. Specifically, the wind speed data and wind direction data at different heights are integrated to form a complete data set. Through statistical analysis methods (such as calculating the correlation coefficient and performing regression analysis, etc.), the correlation between the wind speed gradient and the wind direction change characteristics is explored. Based on the results of the correlation analysis, a mathematical model is constructed to describe the influence of the wind speed gradient on the wind direction change characteristics. Using the constructed model, according to the real-time collected wind speed gradient data, the correction calculation of the wind direction change characteristics is carried out, including the adjustment of the wind direction fluctuation amplitude, wind direction change frequency, and wind direction change rate. Obtaining the wind speed gradient by collecting multiple wind speed data through the wind speed sensor and correcting the wind direction change characteristics according to the wind speed gradient provides more powerful support for the operation and maintenance of the wind turbine, thereby improving the operation efficiency and safety of the wind turbine, and reducing the failure risk and operation and maintenance costs.

[0029] Step S300, establish a load calculation model according to the unit structure parameters and the wind direction change characteristics, and obtain the load analysis result of the target wind turbine.

[0030] Preferably, based on the collected unit structure parameters and the determined wind direction change characteristics, a load calculation model is established. Specifically, the wind force acting on the wind turbine, including tangential force and normal force, etc., is calculated according to the wind direction and wind speed. The structural response of the wind turbine, such as displacement, strain, and stress, etc., is calculated according to the wind force and the unit structure parameters. At the same time, the response of the target wind turbine under dynamic wind loads, such as vibration and fatigue, etc., is considered. Based on these forces, a load calculation model is established using the momentum - blade element theory, and then the load calculation analysis of the target wind turbine is carried out. Specifically, the unit structure parameters and the wind direction change characteristics, etc., are input into the model, and the load state of the wind turbine at different wind directions and wind speeds is calculated according to the model. The calculation results are analyzed and evaluated to understand the stress state and potential risks of the wind turbine under different working conditions, and the load analysis result of the target wind turbine is output.

[0031] Step S300 further includes step S310 of collecting the ambient temperature data and ambient humidity data of the target wind turbine; step S320 of establishing the environmental impact factor of the target wind turbine based on the ambient temperature data and the ambient humidity data; and step S330 of optimizing the load calculation model according to the environmental impact factor and adjusting the load analysis result.

[0032] Preferably, temperature and humidity sensors are installed around or at specific positions of the target wind turbine to collect the data of ambient temperature and ambient humidity in real time. Since ambient temperature and ambient humidity are important factors affecting the performance and load of wind turbines, these ambient temperature and humidity data are usually transmitted and recorded through a sensor network. Then, based on the collected ambient temperature and humidity data, for example, through statistical analysis, machine learning or physical modeling methods, an environmental impact factor is established. Among them, the environmental impact factor is a comprehensive index describing the impact of environmental factors (such as temperature, humidity, etc.) on the performance and load of wind turbines. Specifically, preprocess the collected ambient temperature and humidity data, such as cleaning, denoising and normalization processing, to ensure the accuracy and consistency of the data. Extract features from the preprocessed data that can reflect the changes in ambient temperature and humidity, such as mean value, standard deviation, maximum value, minimum value, etc. Based on the extracted features, construct a mathematical model of the environmental impact factor, and verify the accuracy and reliability of the model by comparing with actual observation data or historical data. Then optimize the load calculation model according to the established environmental impact factor to more accurately reflect the impact of ambient temperature and humidity on the load. Specifically, it may include adjusting relevant parameters in the load calculation model according to the changes in the environmental impact factor to reflect the direct impact of environmental factors on the load; improving the structure of the load calculation model to better adapt to the changes in ambient temperature and humidity. For example, new variables can be introduced or the complexity of the model can be adjusted; finally, use the optimized load calculation model to adjust the original load analysis result. This process may include re-evaluating the load distribution, load peak and load change trend. Through this optimization process, more accurate and reliable load analysis results can be obtained, providing more powerful and precise support for the operation and maintenance of wind turbines. At the same time, it also helps to reduce the failure risk and operation and maintenance costs, and improve the operation efficiency and safety of wind turbines.

[0033] Step S330 further includes step S331 of obtaining the real-time power output parameters of the target wind turbine; step S332 of performing correlation analysis on the real-time power output parameters and the load analysis result to evaluate the operation efficiency of the target wind turbine; and step S333 of generating an operation and maintenance warning signal if the operation efficiency of the target wind turbine decreases.

[0034] Preferably, the real-time power output parameter is an important indicator for measuring the operating status and power generation efficiency of a wind turbine. Install corresponding sensors or data acquisition devices in the control system of the target wind turbine to obtain the real-time power output parameter. These devices can monitor the power output of the wind turbine in real time and convert it into digital signals for recording and analysis. The real-time power output parameter usually includes instantaneous power, average power, power fluctuation, etc., which can reflect the power generation capacity and stability of the wind turbine at different time periods. Then, perform a correlation analysis on the real-time power output parameter and the load analysis results (including key parameters such as blade load, tower load, and main shaft load) to evaluate the operating efficiency of the target wind turbine. Specifically, ensure that the real-time power output parameter and the load analysis results are consistent in time for accurate comparative analysis. Through statistical analysis methods, explore the correlation between the real-time power output parameter and the load analysis results. Based on the results of the correlation analysis, evaluate the operating efficiency of the target wind turbine. If there is a positive correlation between the real-time power output parameter and the load analysis results, and the power output is stable and high, it can be considered that the operating efficiency of the wind turbine is high. On the contrary, if the power output fluctuates greatly or decreases, and the load analysis results also show abnormal stress conditions, it may mean that the operating efficiency of the wind turbine has decreased. If it is found through the correlation analysis that the operating efficiency of the target wind turbine has decreased, generate an operation and maintenance warning signal to remind the operation and maintenance personnel to take measures in a timely manner. Among them, the content of the operation and maintenance warning signal usually includes the warning level. Determine the level of the warning signal (such as low level, medium level, high level, etc.) according to the degree of reduction in operating efficiency and the size of potential risks. Warning reason: Briefly explain the possible reasons for the reduction in operating efficiency, such as blade wear, tower loosening, main shaft failure, etc. Suggested measures: Provide suggested measures for the warning reason, such as shutdown inspection, component replacement, parameter adjustment, etc. Through this process, the operation and maintenance personnel can timely discover and solve problems in the operation of the wind turbine, ensuring the stable operation and efficient power generation of the unit. At the same time, it also helps to reduce the failure risk and operation and maintenance costs, improving the overall economic benefits of the wind farm.

[0035] Step S400, construct a fault warning model. Further, step S400 further includes step S410. The key components of the target wind turbine include blades, towers, and main shafts. Step S420, obtain the historical fault records of the target wind turbine, and determine the blade fault data sample, tower fault data sample, and main shaft fault data sample. Step S430, respectively construct a first fault warning sub-model, a second fault warning sub-model, and a third fault warning sub-model according to the blade fault data sample, tower fault data sample, and main shaft fault data sample. Step S440, synthesize the fault warning model with the first fault warning sub-model, the second fault warning sub-model, and the third fault warning sub-model.

[0036] Preferably, a fault warning model for the target wind turbine is constructed, that is, historical fault data is used to predict and prevent possible future faults. Specifically, the target wind turbine and its key components are identified, including blades, tower and main shaft, which are the most important components of the wind turbine, and their performance and state directly affect the overall operation efficiency and safety of the wind turbine. The historical fault records of the target wind turbine are collected, including information such as the time, location, component type, fault phenomenon, fault cause and maintenance measures of the fault occurrence. After collecting the historical fault records, it is necessary to determine the blade fault data sample, tower fault data sample and main shaft fault data sample respectively according to the fault conditions of the blades, tower and main shaft; according to the determined fault data samples, the first fault warning sub-model (for blades), the second fault warning sub-model (for tower) and the third fault warning sub-model (for main shaft) are constructed respectively. These fault warning sub-models are constructed and trained using machine learning algorithms (such as support vector machine, neural network, decision tree, etc.). Specifically, when constructing the model, the fault data samples are divided into a training set and a test set. The training set is used to train the model so that it can learn the laws and characteristics of fault occurrence; the test set is used to evaluate the performance of the model to ensure its sufficient accuracy and reliability in practical applications; after constructing the fault warning sub-models of the three key components, they are fused and integrated to obtain a comprehensive fault warning model. The fault warning model is applied to the actual operation of the target wind turbine. By real-time monitoring and analyzing the state data of the key components, potential faults can be predicted and early warning signals can be sent in advance.

[0037] Step S430 further includes step S431 of extracting a blade load data sample and a blade fault data sample from the blade fault data sample; step S432 of training to obtain the first fault warning sub-model based on a neural network model with the blade load data sample as the training input and the blade fault data sample as the training output; step S433 of similarly training to obtain the second fault warning sub-model and the third fault warning sub-model.

[0038] Preferably, extract the blade load data samples and blade fault data samples from the blade fault data samples, and train the first fault warning sub-model (for the blade) based on these samples, and similarly train the second fault warning sub-model (for the tower) and the third fault warning sub-model (for the main shaft). Specifically, screen out the fault data related to the blade from the historical fault records of the target wind turbine to form the blade fault data samples, including the specific information of the blade fault, such as the fault type, fault occurrence time, fault location, etc.; extract the corresponding data related to the blade load to form the blade load data samples, including the load status of the blade under the corresponding working conditions, such as tangential force, normal force, bending moment, torque, etc.; construct a warning model based on a neural network model (such as a multi-layer perceptron, convolutional neural network, etc.), use the blade load data samples as the training input, and the blade fault data samples as the training output, train the neural network model, and after completion, use another part of the blade fault data samples that did not participate in the training to verify the model to obtain the first fault warning sub-model for the blade fault warning of the target wind turbine. Similarly, extract the tower fault data samples from the historical fault records, and extract the data samples related to the tower load from the real-time wind power data and the unit structure parameters, and use these data samples to train the second fault warning sub-model (for the tower); extract the main shaft fault data samples from the historical fault records, and extract the data samples related to the main shaft status from the real-time wind power data and the unit structure parameters, and use these data samples to train the third fault warning sub-model (for the main shaft).

[0039] Step S500, if there is an abnormal load change in the load analysis result, generate a fault warning signal according to the fault warning model.

[0040] Preferably, when there is an abnormal load change in the load analysis result, generate a fault warning signal according to the fault warning model. Specifically, monitor the load data of the target wind turbine in real time, including monitoring the load status of key components such as the blades, tower, and main shaft of the target wind turbine under different working conditions, and identify whether there is an abnormal load change according to the preset load threshold, such as a sudden increase, decrease, or fluctuation of the load. Then, use the identified abnormal load change as the input data and input it into the previously trained fault warning model. The fault warning model will perform fault prediction based on the input abnormal load change data, such as predicting the type, occurrence time, and possible impacts of the target wind turbine fault. If the model predicts the existence of potential fault risks, it will automatically generate a fault warning signal, which helps the operation and maintenance personnel to timely discover and handle potential fault risks and ensure the stable operation and safety of the wind turbine.

[0041] Step S500 further includes step S510. If the blade load of the target wind turbine exceeds the first load threshold, the first fault warning model generates a first fault warning signal; step S520. If the tower load of the target wind turbine exceeds the second load threshold, the second fault warning model generates a second fault warning signal; step S530. If the main shaft load of the target wind turbine exceeds the third load threshold, the third fault warning model generates a third fault warning signal.

[0042] Preferably, based on the load analysis result of the target wind turbine, the blade load of the unit is determined. The first load threshold is a load standard value set based on the normal working range of the blade and past experience data to judge whether the blade load is abnormal. When the real-time monitoring value of the blade load exceeds the first load threshold, the first fault warning model will judge whether the abnormality can trigger a fault warning. If it is confirmed that there is a fault risk, the first fault warning model will generate a first fault warning signal; similarly, based on the load analysis result of the target wind turbine, the tower load of the unit is determined. The second load threshold is a load standard value set based on the normal working range of the tower and past experience data to judge whether the tower load is abnormal. When the real-time monitoring value of the tower load exceeds the second load threshold, the second fault warning model will judge whether the abnormality can trigger a fault warning. If it is confirmed that there is a fault risk, the second fault warning model will generate a second fault warning signal; based on the load analysis result of the target wind turbine, the main shaft load of the unit is determined. The third load threshold is a load standard value set based on the normal working range of the main shaft and past experience data to judge whether the main shaft load is abnormal. When the real-time monitoring value of the main shaft load exceeds the third load threshold, the third fault warning model will judge whether the abnormality can trigger a fault warning. If it is confirmed that there is a fault risk, the third fault warning model will generate a third fault warning signal. Through the wind turbine fault warning signal generation mechanism based on the load threshold, it helps the operation and maintenance personnel to timely discover and handle potential fault risks, and ensure the stable operation and safety of the wind turbine.

[0043] The above specific embodiments do not constitute a limitation to the protection scope of the present application. Those skilled in the art should understand that various modifications, combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principle of the present application shall be included within the protection scope of the present application. In some cases, the actions or steps recorded in the present application can be executed in a different order from that in the embodiments and still achieve the desired results. Additionally, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

Claims

1. A fault warning method for a wind turbine combined with load analysis, characterized in that, The method includes: Collecting real-time wind force data and unit structure parameters of the target wind turbine; Analyzing the load impact of the wind direction on the target wind turbine based on the real-time wind force data, and obtaining the wind direction change characteristics of the target wind turbine; Establishing a load calculation model according to the unit structure parameters and the wind direction change characteristics, and obtaining the load analysis result of the target wind turbine; Constructing a fault warning model; If there is an abnormal load change in the load analysis result, generating a fault warning signal according to the fault warning model; The constructing of the fault warning model includes: The key components of the target wind turbine include blades, tower, and main shaft; Obtaining the historical fault records of the target wind turbine, and determining the blade fault data sample, tower fault data sample, and main shaft fault data sample; Respectively constructing a first fault warning sub-model, a second fault warning sub-model, and a third fault warning sub-model according to the blade fault data sample, tower fault data sample, and main shaft fault data sample; Synthesizing the fault warning model with the first fault warning sub-model, the second fault warning sub-model, and the third fault warning sub-model.

2. The fault warning method for a wind turbine unit combining load analysis according to claim 1, wherein The analyzing of the load impact of the wind direction on the target wind turbine includes: Using a wind direction sensor to collect the wind direction data at the location of the target wind turbine in real time; Performing dynamic analysis on the wind direction data to obtain the wind direction change characteristics, wherein the wind direction change characteristics include the wind direction fluctuation amplitude, wind direction change frequency, and wind direction change rate; Identifying the load change when the wind direction changes based on the wind direction change characteristics.

3. The fault warning method for a wind turbine unit combined with load analysis according to claim 1, characterized in that, The method includes: Collecting multiple wind speed data at different heights of the target wind turbine through a wind speed sensor to obtain a wind speed gradient; Correcting the wind direction change characteristics according to the wind speed gradient.

4. The wind turbine fault warning method combining load analysis according to claim 2, wherein, Including: Collecting the ambient temperature data and ambient humidity data of the target wind turbine; Based on the ambient temperature data and the ambient humidity data, establishing an environmental impact factor of the target wind turbine; Optimizing the load calculation model according to the environmental impact factor and adjusting the load analysis result.

5. The fault warning method for a wind turbine unit combined with load analysis according to claim 1, characterized in that, The method includes: Obtaining the real-time power output parameters of the target wind turbine; Performing correlation analysis on the real-time power output parameters and the load analysis result to evaluate the operation efficiency of the target wind turbine; If the operation efficiency of the target wind turbine decreases, generating an operation and maintenance warning signal.

6. The fault warning method for a wind turbine unit combining load analysis according to claim 1, characterized in that, Respectively constructing a first fault warning sub-model, a second fault warning sub-model, and a third fault warning sub-model according to the blade fault data sample, tower fault data sample, and main shaft fault data sample, includes: Extracting the blade load data sample and the blade fault data sample from the blade fault data sample; Using the blade load data sample as the training input and the blade fault data sample as the training output, and training to obtain the first fault warning sub-model based on the neural network model; Similarly, training to obtain the second fault warning sub-model and the third fault warning sub-model.

7. The method for early warning of wind turbine faults combining load analysis according to claim 6, wherein, If there is an abnormal load change in the load analysis result, generating a fault warning signal according to the fault warning model, includes: If the blade load of the target wind turbine exceeds the first load threshold, the first fault warning sub-model generates a first fault warning signal; If the tower load of the target wind turbine exceeds the second load threshold, the second fault warning sub-model generates a second fault warning signal; If the main shaft load of the target wind turbine exceeds the third load threshold, the third fault warning sub-model generates a third fault warning signal.

Images