Method and system for carrying out early warning control on wind power equipment based on GIS (Geographic Information System)

Abstract

The invention provides a method and a system for performing early warning control on wind power equipment based on a GIS (Geographic Information System), and the method comprises the following steps: obtaining point location data of the wind power equipment through the GIS, and classifying the wind power equipment based on the point location data; the method comprises the following steps: acquiring historical meteorological data around each category of wind power equipment based on classification of the wind power equipment, acquiring wind power floating characteristic data of the wind power equipment based on the historical meteorological data, constructing a wind power prediction model of the corresponding category of wind power equipment based on the wind power floating characteristic data, and collecting current meteorological data around each category of wind power equipment, and obtaining wind power floating characteristic data of the corresponding wind power equipment based on the current meteorological data, predicting future wind power data of the wind power equipment in each category based on the wind power prediction model, and controlling pre-steering, braking and early warning of fan blades of the wind power equipment in the corresponding category based on the predicted future wind power data. According to the invention, the problem that traditional wind power equipment is passive and depends on staff to monitor each wind power equipment on site when facing different wind can be solved.

Description

TECHNICAL FIELD

[0001] The present application mainly relates to the field of wind power equipment monitoring, and particularly relates to a method and system for early warning control of wind power equipment based on GIS. BACKGROUND

[0002] With the increasing global energy demand, traditional energy gradually cannot meet the needs of people's life, clean energy, such as wind energy, solar energy, etc. has been widely developed, and the research and development of related technologies have become a hot spot As one of the most promising clean energy, the development and utilization of wind energy has been widely valued in the world. However, with the continuous expansion of the scale and quantity of wind power equipment, the supervision of wind power equipment also continues to increase. The traditional monitoring relies on on-site monitoring of employees, which has high investment cost. At the same time, employees need to check the aging condition of each wind power equipment, which is low in efficiency. Moreover, the traditional wind power equipment relies on the tail end sensor to sense the wind direction to control the wind blade to turn, and when facing strong wind, the wind blade needs to be passively braked by high-speed rotation, which not only wastes part of the wind power resources, but also accelerates the aging of the wind power equipment.

[0003] Therefore, the present application designs a method and system for early warning control of wind power equipment based on GIS to solve the above problems that the traditional wind power equipment is passive and relies on employees to monitor each wind power equipment on site when facing different winds. SUMMARY

[0004] In order to solve the problems of the prior art, the present application combines the prior art and provides a method and system for early warning control of wind power equipment based on GIS to solve the above problems that the traditional wind power equipment is passive and relies on employees to monitor each wind power equipment on site when facing different winds.

[0005] The technical scheme of the present application is as follows: According to one aspect of the present application, a method for early warning control of wind power equipment based on GIS is provided, comprising the following steps: S1, obtaining point data of wind power equipment through GIS, and classifying the wind power equipment based on the point data, wherein the point data includes the location of the wind power equipment and the geographical type of the location; S2, obtaining historical meteorological data around the wind power equipment under each category based on the classification of the wind power equipment, and obtaining wind floating characteristic data of the wind power equipment based on the historical meteorological data, wherein the wind floating characteristic data includes wind adaptation attribute and wind friction attribute; S3, constructing a wind power prediction model of the wind power equipment under the corresponding category based on the wind floating characteristic data, wherein the wind power prediction model is used to calculate future wind power. S4, collect current meteorological data around the wind power equipment under each category, obtain wind floating characteristic data of the corresponding wind power equipment based on the current meteorological data, and predict future wind data of the wind power equipment under each category based on the wind power prediction model; S5, based on the predicted future wind data, control the pre-rotation, braking and early warning of the wind blade of the wind power equipment under the corresponding category.

[0006] Further, in step S1, the location of the wind power equipment is divided by city level, and the geographical types include mountain, plateau, basin, plain and hilly land. When classifying the wind power equipment, first, the geographical type is classified in the first level, and then the location in the same geographical type is classified in the second level.

[0007] Further, the historical meteorological data and the current meteorological data at least include atmospheric pressure around the wind power equipment, temperature and humidity information around the wind power equipment, and wind power information around the wind power equipment.

[0008] Further, the method for obtaining the wind floating characteristic data is as follows: Based on the historical meteorological data or the current meteorological data, the corresponding atmospheric condition, temperature and humidity condition and wind condition are obtained; Set the wind environment coefficient, and obtain the wind adaptation attribute according to the wind environment coefficient, atmospheric condition, temperature and humidity condition and wind condition; Set the wind resistance coefficient, and obtain the wind friction attribute according to the wind resistance coefficient and the wind friction value; Based on the wind adaptation attribute and the wind friction attribute, the wind floating characteristic data is obtained.

[0009] Further, the method for obtaining the wind adaptation attribute is as follows: The wind environment coefficient, air density correction coefficient, atmospheric stability and wind condition are brought into the fluid mechanics software to simulate the pressure field and vortex shedding effect when the wind flows through the object, and then the wind adaptation attribute is calculated, wherein the air density correction coefficient is converted by the temperature and humidity condition, and the atmospheric stability is converted by the atmospheric condition; The method for obtaining the wind friction attribute is as follows: the satellite data is used to analyze the ground texture, and the machine learning is used to estimate the large-scale roughness; The wind floating characteristic is calculated based on the random forest algorithm, combined with the wind adaptation attribute and the wind friction attribute.

[0010] Further, the method for constructing the wind power prediction model is as follows: The historical meteorological data of continuous x time points is obtained as input data; The historical meteorological data of the next continuous x time points is obtained as output data; The data set is constructed combined with the input data and the output data; The NWP model is trained by using a data set and wind floating characteristic data to obtain a wind power prediction model.

[0011] Further, the specific process of model training is as follows: The data set and wind floating characteristic data are preprocessed to obtain a large number of data samples in the form of time series. All data samples are divided into a training set and a test set. A prediction model based on a recurrent neural network is built. The data of the training set is input into the model for training. The data of the test set is input into the trained prediction model for verification. A wind power prediction model is obtained. The future wind power data predicted by the wind power prediction model includes wind direction, movement trajectory, movement speed and intensity information.

[0012] Step S5 specifically includes: The time, direction and intensity data of the wind reaching the device are calculated according to the predicted future wind power. The data is transmitted to the control system of the wind power generator, so as to control the yaw motor and the variable pitch control system to adjust the direction and angle of the wind blade. If the intensity of the wind exceeds the safety threshold of the wind power device, the braking device will be controlled in advance to brake the wind blade, and the braking information of this time will be uploaded to the GIS for storage and early warning.

[0013] Further, the wind power device is built-in with a simulation aging curve model based on life data to the GIS before being shipped and running. The aging degree of the wind power device is monitored in real time based on the simulation aging curve model each time the wind power device is running. If the wind power device reaches a preset aging threshold, a warning prompt is given on the GIS.

[0014] According to another aspect of the present application, a system for early warning control of wind power devices based on GIS is provided, which comprises: A data acquisition unit configured to obtain point data of the wind power device through the GIS; A classification unit configured to classify the wind power device based on the point data, wherein the point data includes the location of the wind power device and the geographical type of the location; An analysis unit configured to obtain historical meteorological data around the wind power device under each category based on the classification of the wind power device, and obtain wind floating characteristic data of the wind power device based on the historical meteorological data, wherein the wind floating characteristic data includes wind adaptation properties and wind friction properties; A construction unit configured to construct a wind power prediction model of the wind power device under the corresponding category based on the wind floating characteristic data, wherein the wind power prediction model is used to calculate future wind power; The prediction unit is configured to collect current meteorological data around the wind power equipment in each category, obtain wind floating characteristic data of the corresponding wind power equipment based on the current meteorological data, and predict future wind data of the wind power equipment in each category based on a wind power prediction model. The control unit is configured to control the pre-rotation, braking and early warning of the wind blade of the wind power equipment in the corresponding category based on the predicted future wind data. The monitoring unit is configured to monitor the aging degree of the wind power equipment in real time based on the simulation aging curve model, and if the wind power equipment reaches a preset aging threshold, a warning prompt is given on the GIS.

[0015] The present application has the following advantages: The present application combines wind power equipment with GIS, first obtains historical meteorological data around the wind power equipment, obtains wind floating characteristics around the wind power equipment, and then constructs a wind power prediction model, detects meteorological data around the wind power equipment, calculates future wind power according to the wind power prediction model, controls the pre-rotation and braking of the wind blade of the wind power equipment, so that the wind power equipment is no longer passive when facing different winds, not only increases the conversion of wind power resources, but also slows down the aging of the wind power equipment.

[0016] The present application displays the aging degree of the wind power equipment by using GIS, so that the staff can more conveniently watch the state of each wind power equipment, detects the aging degree of the wind power equipment in real time, modifies the simulation aging curve model according to the aging degree, so that the staff can accurately see the remaining life of each wind power equipment, and at the same time, when the wind power equipment reaches a preset aging threshold, the GIS also gives a warning prompt, providing convenience for the staff to detect the wind power equipment. BRIEF DESCRIPTION OF DRAWINGS

[0017] Figure 1 The wind power equipment control step flow chart of the present application; Figure 2 The wind power equipment aging warning step flow chart of the present application; Figure 3 The wind power equipment control system principle diagram of the present application; Figure 4 The wind power equipment aging warning system principle diagram of the present application. DETAILED DESCRIPTION

[0018] The present application will be further described in conjunction with the drawings and specific embodiments. It should be understood that these embodiments are only used to illustrate the present application and not to limit the scope of the present application. In addition, it should be understood that after reading the content taught by the present application, those skilled in the art can make various modifications or modifications to the present application, and these equivalent forms also fall within the scope defined by the present application. Example 1

[0019] The embodiment provides a method for early warning control of wind power equipment based on GIS.

[0020] Please refer to Figure 1 and Figure 2 The method provided by the embodiment mainly comprises the following steps: Step S1: Obtain point data of each wind power equipment through GIS, and classify each wind power equipment according to different geographical types.

[0021] In this step, the longitude, latitude, altitude and other point data of each device can be collected through the GPS module of the wind power equipment, and uploaded to the GIS system in real time through the 4G / 5G module; at the same time, GIS calls the built-in terrain layer data (such as digital elevation model DEM) to obtain the geographical type label of each point.

[0022] As a preferred scheme of the embodiment, the classification of GIS for each wind power equipment is according to the difference of geographical types (including mountain, plateau, basin, plain and hilly) for one-level classification, and the same geographical type is aggregated and saved together.

[0023] When classifying each wind power equipment, GIS will also classify the wind power equipment according to the difference of the city where the wind power equipment is located, for example, wind power equipment A and B are located on the plain in x city, and wind power equipment C is located on the plain in y city, then GIS aggregates A, B and C according to the plain terrain and saves them together, and then separates and saves A, B and C according to the difference of the city.

[0024] Through two-level refinement, the wind power equipment can be accurately classified, so that the subsequent wind prediction is more targeted, and the prediction data is more accurate.

[0025] Step S2: Obtain historical meteorological data around each wind power equipment in each category, and then obtain the wind floating characteristics of the corresponding category based on the historical meteorological data of the category.

[0026] The historical meteorological data includes atmospheric pressure around the wind power equipment, temperature and humidity information around the wind power equipment, and current wind information around the wind power equipment.

[0027] The specific process of obtaining wind floating characteristics through meteorological data is as follows: According to the historical meteorological data or the current meteorological data, the atmospheric condition, the temperature and humidity condition and the wind condition are obtained; the wind environment coefficient is set, the wind adaptation attribute is obtained according to the wind environment coefficient, the atmospheric condition, the temperature and humidity condition and the wind condition; the wind resistance coefficient is set, the wind friction attribute is obtained according to the wind resistance coefficient and the historical wind friction value; finally, the wind floating characteristic data is obtained according to the wind adaptation attribute and the wind friction attribute.

[0028] In the embodiment, a specific acquisition method of wind floating characteristic data is provided, and the process is as follows: First, set v (wind speed) as 5.48, 7.14, 9.12 and 10.52. Then, according to the formula P t =2000×( )3, the P t (power) is calculated as 243.73, 672.54, 1268.32 and 1785.43; it should be noted that when v<3.5 and v>25.0, P t is 0, and when v is between 12.0 and 25.0 (including 12.0 and 25.0), P t is 2000. Then, according to the formula ΔP avg = , the average power change intensity in the unit time is derived, and after the above data is brought in, ΔP avg =551.36 kW is obtained. Then, the maximum power change is calculated, according to the formula ΔP max = |P t -P t-1 |, it is calculated that when t is between 8 and 9, ΔP max is the maximum, which is 1083.77kW. Then, the power standard deviation is calculated, according to the formula , the power standard deviation is calculated as 805.70kW, wherein μP= . Then, the fluctuation frequency is calculated, according to the formula , the frequency is calculated as 100%, it should be noted that the threshold value=0.1*2000=200kW. Finally, the proportion of zero power time is calculated, according to the formula , the proportion is calculated as 0%.

[0029] In a preferred scheme provided in the embodiment, the specific acquisition method of wind adaptation property is: the wind environment coefficient, the air density correction coefficient and the atmospheric stability and the historical wind conditions are brought into the fluid mechanics software (such as ANSYS Fluent) to simulate the pressure field and vortex shedding effect when the wind flows through the object, and then the wind adaptation property is calculated. The air density correction coefficient is obtained by converting the temperature and humidity conditions, and the atmospheric stability is obtained by converting the atmospheric conditions.

[0030] In a preferred scheme provided in the embodiment, the specific method for obtaining the wind friction attribute is to analyze the surface texture by satellite (such as SAR) data, and estimate the large-scale roughness by machine learning.

[0031] It should be noted that the specific way of estimating the large-scale roughness by machine learning is: Data collection: Obtain directly measured wind stress, 10-meter height wind speed (U 10 ), wave height , wave period through the ocean buoy network, and then collect sea surface wind speed / direction, significant wave height and sea surface temperature through satellite remote sensing scatterometer, altimeter and radiometer, and finally use ERA5 to provide atmospheric stability and boundary layer height, and finally use the drag coefficient measured by the flux tower; Preprocessing: first align the satellite data with the buoy observation (± 30 minutes, ± 25km grid), and then remove outliers (such as remove U 10 < 2m / s and U 10 > 40m / s data and satellite observation data affected by rain); Feature construction: first calculate the air density by the formula ρ a = , P is the air pressure, R d is the dry air gas constant, and T v is the virtual temperature (considering the humidity effect), and then calculate the wave age according to the formula wave age= , where c p is the wave peak phase velocity (which can be estimated from the wave period T p : c p = ), and then calculate the sea-air temperature difference according to the formula ΔT = SST - Tair, where SST is the sea surface temperature, and then calculate the atmospheric stability parameter according to the formula , where u is the friction velocity, κ is the Karman constant, g is the gravitational acceleration, is the sensible heat flux (if the turbulent flux data is missing, the bulk formula can be used to estimate the stability), and finally the input features are standardized or normalized; Feature selection: in the convolutional neural network, use the random forest ensemble model, combine support vector regression and physical information neural network, and train the data set. Specifically, divide the data set (note to divide in time sequence), divide into training set (70%), validation set (15%), and test set (15%), then use the validation set to optimize the hyperparameters, use the loss function mean square error, and finally minimize the prediction error of the wind friction attribute; Verification and evaluation: Time series cross-validation is used, and the results are compared with those of a classic parameterization scheme (such as the COARE 3.0 algorithm), followed by evaluation based on root mean square error (RMSE), mean absolute error (MAE), coefficient of determination (R²), and relative error (with a focus on high wind speed segments).

[0032] In a preferred scheme provided in the embodiment, the specific method of wind floating characteristics is to calculate the wind floating characteristics based on the random forest algorithm combined with wind adaptation attributes and wind friction attributes. The random forest method is an ensemble learning algorithm that improves the accuracy and stability of prediction or classification by constructing multiple decision trees and integrating their results. In this embodiment, the wind adaptation attributes (reflecting the adaptability of the wind environment and the device, calculated based on atmospheric conditions, temperature and humidity, wind power, etc. through fluid mechanics simulation) and wind friction attributes (reflecting the interaction between wind and the ground, calculated based on surface roughness and wind resistance coefficient) are used as input features to construct multiple decision trees to learn these features. Each tree is trained based on randomly sampled samples and features, and finally outputs the comprehensive results through majority voting or averaging to obtain wind floating characteristics that can reflect the variation of wind power with the environment, providing key basis for subsequent construction of wind power prediction model. This method can effectively handle nonlinear relationships and high-dimensional data, improve the accuracy and robustness of wind floating characteristics calculation, and adapt to complex and variable wind environment characteristics in different geographical types.

[0033] Step S3: Construct a wind power prediction model specific to the category based on the wind floating characteristics of each category.

[0034] The specific construction process of the wind power prediction model is as follows: Obtain historical meteorological data at consecutive x time points as input data; Obtain the next consecutive x time electric historical meteorological data as output data; Combine the input data and output data to construct a data set; Use the data set and wind floating characteristics to train the NWP model to obtain the wind power prediction model.

[0035] In a preferred scheme provided in the embodiment, when constructing the data set, the input data and output data are first cleaned to exclude outliers, then smoothed and reduced in dimension, followed by time series feature extraction and encoding, and finally using PyG to construct a graph neural network data set to obtain the data set.

[0036] In a preferred scheme provided in the embodiment, the specific process of model training is: pre-processing the data set and wind floating characteristic data to obtain a large number of data samples in the form of time series, dividing all data samples into a training set and a test set, building a prediction model based on a recurrent neural network (such as RNN), taking the data of the training set as the input of the model, training the network model, inputting the data of the test set into the trained prediction model for verification, and obtaining a wind prediction model. After inputting the wind floating characteristic data, the obtained wind prediction model can accurately output the predicted future wind data.

[0037] It should be noted that the prediction model includes three structures of an input layer, a recurrent layer and an output layer. The input layer is a fixed-length sliding window, which contains observation data of T time steps in the past. The input of each time step T is a feature vector, which includes one or more of the following feature data: historical wind speed (possibly at multiple heights), historical wind direction, historical temperature, historical air pressure, historical humidity, historical turbulence intensity, historical power output (if power prediction), and other related meteorological or site features (such as hour, week, month, etc.).

[0038] The recurrent layer is a multi-layer stacked LSTM unit layer, each LSTM unit layer includes a gating mechanism (input gate, forget gate, output gate) and a cell state. When data enters the recurrent layer, the input gate decides which new information about the current input is stored in the cell state, the forget gate decides which information is discarded from the cell state, and the output gate decides what to output. The cell state is the cell state of the previous time step, which is the "memory" of the network.

[0039] The output layer is one or more fully connected layers, which are used to map the complex patterns learned by the recurrent layer to the final wind prediction value. The first layer is one or more hidden layers with activation functions (such as ReLU), which are used to further process features. The last layer is a linearly activated fully connected layer, which is used for single-step prediction / multi-step prediction (progressive method) tasks. The final structure of the prediction is the wind prediction value.

[0040] Step S4: Collect current meteorological data around each wind power equipment of each category, and obtain wind floating characteristic data based on the current meteorological data, and input the wind floating characteristic data into the wind power prediction model of the category to predict future wind. The predicted future wind specifically refers to the movement direction, movement trajectory, movement speed and intensity information of the wind in the future short term.

[0041] Step S5: According to the predicted future wind, the wind power equipment of the category is controlled for early warning.

[0042] The early warning control specifically comprises: calculating time, direction and intensity data of wind reaching the device according to the predicted future wind; setting a program according to the calculated data and transmitting the program signal to the control system of the wind turbine, so as to control the yaw motor and the variable pitch control system to control the wind blade to adjust the direction and angle; if the intensity of the wind exceeds the safety threshold of the wind power device, the brake device will be controlled in advance to brake the wind blade, and the braking information of this time will be uploaded to the GIS for storage.

[0043] In this embodiment, the GIS first acquires the point data of each wind power device, then classifies the wind power devices according to different geographic types, then simultaneously acquires the historical meteorological data around all wind power devices in each category, then acquires the wind floating characteristics corresponding to the category according to the historical meteorological data of the category, then constructs a dedicated wind prediction model according to the wind floating characteristics of each category, then the GIS collects the meteorological data around all wind power devices in each category again, and brings the meteorological data into the wind prediction model corresponding to the category, predicts the future wind, and finally controls the wind power device to turn in advance, brake and give early warning when facing different winds. When facing different winds, the wind power device is no longer passive, not only increases the conversion of wind power resources, but also slows down the aging of the wind power device.

[0044] In a preferred scheme provided in this embodiment, the wind power device can also be warned of aging. Specifically, the wind power device will record and upload a simulation aging curve model based on life data to the GIS before it is shipped and runs. The aging degree of the wind power device is monitored each time it runs, the simulation aging curve model is modified based on the aging degree, and if the wind power device reaches a preset aging threshold, a warning prompt will be given on the GIS. When the wind power device is running, the GIS also monitors and records the aging degree in real time and modifies the simulation aging curve model based on the aging degree. At the same time, if the wind power device reaches the preset aging threshold, a prompt will also be given on the GIS so that employees can quickly find the wind power device.

[0045] It should be noted that the simulation aging curve model describes the decay law of device performance over time through the residual health function, and its application formula is: Residual health ; wherein, t is the cumulative running time of the device (in hours); β is a shape parameter for controlling the shape of the curve. Specifically, when β < 1, it represents the early failure period, at which time the failure rate decreases; when β = 1: it represents the random failure period, at which time the failure rate is constant, which is equivalent to an exponential distribution; when β > 1, it represents the wear-out failure period, at which time the failure rate increases; η is a scale parameter representing the characteristic lifetime. When t = η + γ, the health rate drops to approximately 36.8%. γ is a position parameter, representing the minimum fault-free lifetime threshold (usually γ≥0).

[0046] It should be noted that the remaining health H(t) mentioned above is derived from the power-law characteristic of the failure rate through the integral reliability equation. Specifically, based on the power-law relationship between the aging failure rate λ(t) of mechanical / electrical components and time, we can first derive: Failure rate Among them, when β>1, the failure rate increases over time; Since the reliability function R(t) and the failure rate λ(t) satisfy a differential equation, we can conclude that: λ(t)= lnR(t), where reliability R(t) represents the probability that the device will not fail within time t; Substituting the failure rate function into the equation and integrating it (with the upper limit of integration being γ and the lower limit being t), we obtain: The solution is: Then, the reliability R(t) is converted into a health status in percentage form, i.e.: H(t) = R(t) × 100%, thus obtaining the remaining health H(t). Example 2

[0047] This embodiment provides a system for early warning and control of wind power equipment based on GIS.

[0048] Please see Figure 3 , Figure 4As shown, the system comprises a data acquisition unit, a classification unit, an analysis unit, a construction unit, a prediction unit, a control unit and a monitoring unit, the data acquisition unit is used to acquire point data of each wind power equipment, historical meteorological data around each wind power equipment and meteorological data around each wind power equipment, the classification unit is used to classify each wind power equipment according to different geographical types, the analysis unit is used to obtain wind floating characteristics according to historical meteorological data, the construction unit is used to construct a wind power prediction model according to wind floating characteristics, the prediction unit is used to substitute meteorological data into the wind power prediction model to predict future wind power, the control unit is used to control the wind power equipment according to the predicted future wind power, the monitoring unit is used to monitor the aging degree of the wind power equipment, and the simulation aging curve model is modified according to the aging degree, and if the wind power equipment reaches the preset aging threshold, a warning prompt is given on GIS. In this embodiment, the data acquisition unit acquires the point data of each wind power equipment, the historical meteorological data around each wind power equipment and the meteorological data around each wind power equipment, and transmits the point data of each wind power equipment to the classification unit, transmits the historical meteorological data around each wind power equipment to the analysis unit, and transmits the meteorological data around each wind power equipment to the prediction unit. After receiving the point data of each wind power equipment, the classification unit classifies each wind power equipment according to different geographical types of each wind power equipment. After receiving the historical meteorological data around each wind power equipment, the analysis unit analyzes the historical meteorological data around each wind power equipment according to different wind power equipment categories to obtain wind floating characteristics of the same type of wind power equipment, and the construction unit constructs a wind power prediction model of the same type of wind power equipment according to the wind floating characteristics of the same type of wind power equipment. After receiving the meteorological data around each wind power equipment, the prediction unit substitutes the meteorological data around each wind power equipment into the corresponding wind power prediction model to predict the future wind power of the same type of wind power equipment, and then the control unit controls the wind power equipment. At the same time, the detection device also detects the aging degree of the wind power equipment in real time, modifies the simulation aging model, and prompts the wind power equipment that has reached the aging threshold on GIS, thereby providing convenience for employees to detect the wind power equipment.

Claims

1. A method for early warning and control of wind power equipment based on GIS, characterized in that, Includes the following steps: S1. Obtain location data of wind power equipment through GIS, and classify the wind power equipment based on the location data, wherein the location data includes the location of the wind power equipment and the geographical type of the location; S2. Based on the classification of wind power equipment, obtain historical meteorological data around the wind power equipment under each category, and obtain wind force floating characteristic data of wind power equipment based on the historical meteorological data, wherein the wind force floating characteristic data includes wind adaptability attribute and wind friction attribute. S3. Based on the wind force fluctuation characteristic data, construct a wind force prediction model for wind power equipment under the corresponding category. The wind force prediction model is used to calculate future wind force. S4. Collect current meteorological data around wind power equipment of each category, obtain wind force fluctuation characteristic data of the corresponding wind power equipment based on the current meteorological data, and predict the future wind force data of wind power equipment of each category based on the wind force prediction model. S5. Based on predicted future wind data, control the blades of wind turbines in the corresponding category to pre-steer, brake, and provide early warnings.

2. The method for early warning and control of wind power equipment based on GIS according to claim 1, characterized in that, In step S1, the location of wind power equipment is divided by city-level units. Geographical types include mountains, plateaus, basins, plains and hills. When classifying wind power equipment, the first-level classification is based on geographical type, and the second-level classification is based on the location within the same geographical type.

3. The method for early warning and control of wind power equipment based on GIS according to claim 1, characterized in that, The historical and current meteorological data include at least the atmospheric pressure around the wind power equipment, the temperature and humidity around the wind power equipment, and the wind force around the wind power equipment.

4. The method for early warning and control of wind power equipment based on GIS according to claim 3, characterized in that, The method for obtaining the wind-driven floating characteristic data is as follows: Based on historical or current meteorological data, obtain the corresponding atmospheric conditions, temperature and humidity conditions, and wind conditions. Set the wind environment coefficient, and obtain the wind adaptability attributes based on the wind environment coefficient, atmospheric conditions, temperature and humidity conditions, and wind conditions; Set the drag coefficient, and obtain the wind friction properties based on the drag coefficient and the wind friction value; Based on wind adaptation and wind friction properties, wind-driven floating characteristic data are obtained.

5. The method for early warning and control of wind power equipment based on GIS according to claim 4, characterized in that, The method for obtaining the wind adaptability attribute is as follows: The wind environment coefficient, air density correction coefficient, atmospheric stability, and wind conditions are input into fluid dynamics software to simulate the pressure field and vortex shedding effect when wind flows over an object, and then the wind adaptability attributes are calculated. The air density correction coefficient is converted through temperature and humidity conditions, and the atmospheric stability is converted through atmospheric conditions. The method for obtaining wind friction properties is as follows: analyze the surface texture using satellite data and combine it with machine learning to estimate the roughness over a large area; The wind-induced floating characteristics are calculated based on the random forest algorithm, combined with wind adaptation and wind friction properties.

6. The method for early warning and control of wind power equipment based on GIS according to claim 1, characterized in that, The method for constructing the wind prediction model is as follows: Use historical meteorological data for x consecutive time points as input data; The output data is the historical meteorological data for the next x consecutive time points. A dataset is constructed by combining the input and output data; The NWP model was trained using the dataset and wind-induced floating characteristics data to obtain a wind prediction model.

7. The method for early warning and control of wind power equipment based on GIS according to claim 6, characterized in that, The specific process of model training is as follows: The dataset and wind fluctuation characteristic data are preprocessed to obtain a large number of data samples in time series form. All data samples are divided into training set and test set. A prediction model based on recurrent neural network is built. The data in the training set is used as the model input to train the network model. The data in the test set is input into the trained prediction model to verify it, and the wind prediction model is obtained. The future wind data predicted by the wind prediction model includes the wind's direction of movement, trajectory, speed and intensity information.

8. The method for early warning and control of wind power equipment based on GIS according to claim 1, characterized in that, Step S5 specifically includes: The system calculates the time, direction, and intensity of the wind reaching the equipment based on the predicted future wind force. The data is then transmitted to the wind turbine's control system, which in turn controls the yaw motor and pitch control system to adjust the direction and angle of the wind blades. If the wind intensity exceeds the safety threshold of the wind turbine, the system will control the braking device to brake the wind blades in advance, and the braking information will be uploaded to the GIS for storage and an early warning will be issued.

9. The method for early warning and control of wind power equipment based on GIS according to claim 1, characterized in that, Before the wind power equipment is put into operation, a simulation aging curve model based on life data is built into the GIS. During each operation of the wind power equipment, the aging degree of the wind power equipment is monitored in real time based on the simulation aging curve model. If the wind power equipment reaches the preset aging threshold, an early warning will be issued on the GIS.

10. A system for early warning and control of wind power equipment based on GIS, characterized in that, The system includes: The data acquisition unit is configured to acquire location data of wind power equipment via GIS. A classification unit is configured to classify the wind power equipment based on location data, wherein the location data includes the location of the wind power equipment and the geographical type of the location; The analysis unit is configured to acquire historical meteorological data around wind power equipment under each category based on the classification of wind power equipment, and acquire wind force floating characteristic data of wind power equipment based on the historical meteorological data, wherein the wind force floating characteristic data includes wind adaptability attributes and wind friction attributes. The construction unit is configured to construct a wind prediction model for wind power equipment under the corresponding category based on the wind fluctuation characteristic data. The wind prediction model is used to calculate future wind force. The prediction unit is configured to collect current meteorological data around wind power equipment of various categories, obtain wind force fluctuation characteristic data of the corresponding wind power equipment based on the current meteorological data, and predict future wind force data of wind power equipment of various categories based on the wind force prediction model. The control unit is configured to control the blades of wind turbines of the corresponding category to pre-steering, braking, and warning based on predicted future wind data. The monitoring unit is configured to monitor the aging degree of wind power equipment in real time based on a simulated aging curve model. If the wind power equipment reaches the preset aging threshold, an early warning will be issued on the GIS.

Images