Wind farm disaster early warning method, device, computer equipment and storage medium

Abstract

The present invention discloses a wind farm disaster early warning method, device, computer equipment and storage medium, the method comprising: obtaining meteorological element data, weather forecast data, grid forecast data and terrain topology data; fusing meteorological element data into terrain topology data, and interpolating the fused terrain topology data into grid forecast data to obtain a grid element data set; extracting events from weather forecast data to obtain event variables, performing statistics on grid element data sets to obtain diagnostic information, and matching event variables with diagnostic information to generate disaster prediction results for early warning. The present invention collects diagnostic information from integrated data sets, extracts event variables from weather forecast data, and matches event variables with diagnostic information, so that meteorological element data, weather forecast data, grid forecast data and terrain topology data can be combined to generate disaster prediction results, reduce the impact of the environment on disaster prediction results, and improve the effective time and accuracy of early warning.

Description

Technical Field

[0001] The present invention relates to the technical field of wind farm disaster prediction, and in particular to a wind farm disaster early warning method, device, computer equipment and storage medium. Background Art

[0002] Wind power generation has gradually become an important form of new energy technology with its advantages of low pollution and high safety, and has been developed more and more widely. Wind power generation relies on wind farms. Some wind farms are built in remote areas with high altitudes and sparse populations. In these areas, it is easy for disastrous weather to damage the wind power generation equipment in the wind farms, resulting in major accidents and economic losses. Therefore, it is necessary to warn of possible meteorological disasters in wind farms, so as to shut down the wind turbines in time or adjust the operation mode of the wind turbines, and reinforce and protect various equipment and facilities in the wind farms to ensure the safe operation of the entire wind farm.

[0003] There are currently two main methods for early warning of meteorological disasters in wind farms. The first method is to cooperate with the meteorological department to use meteorological satellite images and local ground meteorological stations and other equipment to monitor the environment of wind farms in order to predict the occurrence of meteorological disasters. The second method is to install meteorological instruments on the ground or on high towers within the jurisdiction of the wind farm to collect real-time meteorological data, and use modern meteorology to conduct in-depth analysis of the real-time meteorological data. The analysis system then makes early warning decisions based on changes in real-time meteorological data. However, the two existing early warning methods generally have some defects. For example, the defect of the first early warning method is that the effective warning time is too short, which is not conducive to the arrangement of disaster prevention work in wind farms. In addition, in sparsely populated areas, the number of ground meteorological stations is small. When the direction of the weather system is inconsistent with the location of the ground meteorological station and the location of the wind farm, accidents that should be warned but are not warned are prone to occur; the defect of the second early warning method is that it is necessary to form an additional professional meteorological team to ensure the reliability of the data and the normal operation of the equipment, which requires high additional costs (equipment construction, labor, operating power consumption, etc.). At the same time, from the perspective of modern meteorological methods, the spatial representativeness of meteorological instruments installed in wind farms is poor, and only a certain degree of short-term warning can be made. However, when faced with weather disasters such as thunderstorms and hail that develop at a fast speed, the warning efficiency is low, and it does not have the ability to warn of weather disasters such as cold waves, frost, fog, and strong winds caused by larger-scale weather systems. Summary of the invention

[0004] The embodiments of the present invention provide a wind farm disaster warning method, device, computer equipment and storage medium, which aim to reduce the impact of environmental factors on disaster prediction results, improve the warning effectiveness time and accuracy of disaster prediction results, and reduce the warning cost of wind farm disasters.

[0005] In a first aspect, an embodiment of the present invention provides a wind farm disaster early warning method, comprising:

[0006] Obtain meteorological element data, weather forecast data, grid forecast data and terrain topology data;

[0007] Fusing the meteorological element data into the terrain topology data, and interpolating the fused terrain topology data into the grid forecast data to obtain a grid element data set;

[0008] Event variables are obtained by extracting events from the weather forecast data, diagnostic information is obtained by performing statistics on the grid element data set, and the event variables and diagnostic information are matched to generate disaster prediction results and issue warnings.

[0009] In a second aspect, an embodiment of the present invention provides a wind farm disaster warning device, comprising:

[0010] A data acquisition unit, used to acquire meteorological element data, weather forecast data, grid forecast data and terrain topology data;

[0011] A data integration unit, used for fusing the meteorological element data into the terrain topology data, and interpolating the fused terrain topology data into the grid forecast data to obtain a grid element data set;

[0012] The disaster prediction unit is used to extract events from the weather forecast data to obtain event variables, perform statistics on the grid element data set to obtain diagnostic information, and match the event variables with the diagnostic information to generate disaster prediction results and issue warnings.

[0013] In a third aspect, an embodiment of the present invention provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the wind farm disaster warning method as described in the first aspect is implemented.

[0014] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the wind farm disaster warning method as described in the first aspect is implemented.

[0015] The embodiment of the present invention discloses a wind farm disaster warning method, device, computer equipment and storage medium. The method includes: obtaining meteorological element data, weather forecast data, grid forecast data and terrain topology data; fusing the meteorological element data into the terrain topology data, and interpolating the fused terrain topology data into the grid forecast data to obtain a grid element data set; extracting events from the weather forecast data to obtain event variables, performing statistics on the grid element data set to obtain diagnostic information, and matching the event variables with the diagnostic information to generate disaster prediction results and issue warnings. The embodiment of the present invention first obtains a grid element data set by integrating meteorological element data, grid forecast data and terrain topology data, then extracts events from weather forecast data to obtain event variables, then generates diagnostic information based on the grid element data set, and finally matches the event variables with the diagnostic information, and generates disaster prediction results according to the matching results for warning, so that disaster prediction results can be generated in combination with meteorological element data, weather forecast data, grid forecast data and terrain topology data, thereby reducing the impact of environmental factors on disaster prediction results, improving the effective warning time and accuracy of disaster prediction results, and reducing the warning cost of wind farm disasters. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying any creative work.

[0017] Figure 1 A schematic diagram of a flow chart of a wind farm disaster early warning method provided by an embodiment of the present invention;

[0018] Figure 2 A schematic diagram of a sub-process of a wind farm disaster early warning method provided by an embodiment of the present invention;

[0019] Figure 3 A schematic diagram of another sub-process of a wind farm disaster early warning method provided by an embodiment of the present invention;

[0020] Figure 4 A system block diagram of a wind farm disaster early warning method provided by an embodiment of the present invention;

[0021] Figure 5 Another system block diagram of a wind farm disaster early warning method provided by an embodiment of the present invention;

[0022] Figure 6 A schematic block diagram of a wind farm disaster warning device provided by an embodiment of the present invention;

[0023] Figure 7 A sub-schematic block diagram of a wind farm disaster warning device provided by an embodiment of the present invention;

[0024] Figure 8 Another sub-schematic block diagram of a wind farm disaster warning device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0025] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0026] It should be understood that when used in this specification and the appended claims, the terms "include" and "comprises" indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof.

[0027] It should also be understood that the terms used in this specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the specification of the present invention and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0028] It should be further understood that the term "and / or" used in the present description and the appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0029] See below Figure 1 , Figure 1 A schematic flow chart of a wind farm disaster early warning method provided by an embodiment of the present invention specifically includes: steps S101 to S103.

[0030] S101, obtaining meteorological element data, weather forecast data, grid forecast data and terrain topology data;

[0031] S102, fusing the meteorological element data into the terrain topology data, and interpolating the fused terrain topology data into the grid forecast data to obtain a grid element data set;

[0032] S103, extracting events from the weather forecast data to obtain event variables, performing statistics on the grid element data set to obtain diagnostic information, and matching the event variables with the diagnostic information to generate disaster prediction results and issue warnings.

[0033] In this embodiment, meteorological element data, weather forecast data, grid forecast data and terrain topology data are first obtained, and then the meteorological element data are integrated into the terrain topology data, and then the integrated terrain topology data is interpolated into the grid forecast data, thereby obtaining a grid element data set, and then event extraction is performed based on the weather forecast data to obtain event variables, and diagnostic information is generated based on the grid element data set, and finally the event variables and diagnostic information are matched, and a disaster prediction result is generated according to the matching result for early warning. In this way, the meteorological element data, weather forecast data, grid forecast data and terrain topology data can be combined to generate a disaster prediction result, so that the generated disaster prediction result is less affected by environmental factors, while also improving the warning effectiveness time and accuracy of the disaster prediction result, and reducing the early warning cost of wind farm disasters.

[0034] In a specific embodiment, the grid forecast data and terrain topology data can be obtained from the information or data provided by the atmospheric research institutions and the geographical research institutions, respectively, and the meteorological element data and weather forecast data can be obtained from the data cloud service provided by the meteorological big data company or the meteorological department. Among them, the weather forecast data can include the weather bulletin of the location of the wind farm. The weather bulletin specifically refers to the short text communication record about the local weather provided by the meteorological department. The weather bulletin contains the meteorological disasters that have occurred in the area and the weather conditions that need to be paid attention to in the short term, for example: "It rained in a certain district of a certain city at a certain time on a certain day of a certain month, and the rainfall was Y mm", "It is expected that there will be a strong wind and cooling process in a certain district of a certain city within the next two days", etc. The meteorological element data can specifically be the temperature, water vapor content, radial wind, zonal wind, air pressure and other data of the area corresponding to the wind farm. The terrain topology data can include parameters such as soil type, land utilization rate, soil temperature, vegetation number, albedo, etc.

[0035] In one embodiment, in combination Figure 2 As shown, the step S102 includes steps S201 to S203.

[0036] S201, mapping the grid forecast data from the orthogonal coordinate system to the terrain following coordinate system to obtain the grid forecast data in the terrain following coordinate system;

[0037] S202, obtaining air pressure values ​​at different heights of the wind farm according to the meteorological element data, and substituting the air pressure values ​​into the terrain topology data;

[0038] S203, interpolating the terrain topology data into which the air pressure value is substituted into the grid forecast data in the terrain following coordinate system through a triangulation method to obtain the grid element data set.

[0039] In this embodiment, the grid forecast data in the orthogonal coordinate system is first mapped to the grid forecast data in the terrain following coordinate system, and then the air pressure values ​​at different altitudes in the corresponding area of ​​the wind farm are obtained based on the meteorological element data, and then the obtained air pressure values ​​are substituted into the terrain topology data. Finally, the triangulation method is used for interpolation, and the terrain topology data after the air pressure value is substituted is interpolated into the grid forecast data in the terrain following coordinate system, so as to obtain the grid element data set.

[0040] In specific application scenarios, the grid forecast data obtained from atmospheric research institutions or other data channels generally have coarse grid points and are not accurate enough to be used to forecast some wind farm disasters (such as hail, lightning strikes, frost, etc.). Therefore, it is necessary to add high-resolution terrain topology data to obtain more refined grid forecast data. The formats of terrain topology data and grid forecast data are different, so coordinate conversion is required to map the grid forecast data from the orthogonal coordinate system to the terrain following coordinate system before the terrain topology data can be added to the grid forecast data.

[0041] The terrain following coordinate system, also known as the σ coordinate system, is proposed to study the influence of terrain on atmospheric movement. The σ coordinate system uses the undulating ground surface as an isosurface of the coordinate system, which can divide the same number of layers equally at each position on the ground surface, and can fit the terrain well on the boundary. In specific application scenarios, the terrain around the wind farm may be undulating, resulting in a difference in the ground altitude around the wind farm. In this case, in the land surface model simulation based on fluid dynamics, when the Navier-Stokes equation is vertically averaged, the horizontal scale will be greater than the vertical scale, which makes the value of the vertical integration step in the orthogonal coordinate system always unable to meet the calculation requirements, and the grid near the ground will have jagged edges, which is difficult to fit the boundary, so that the simulation accuracy near the ground is affected. Therefore, by mapping the grid forecast data from the orthogonal coordinate system to the terrain following coordinate system, the same number of layers can be divided at each horizontal position of the grid forecast data, and the boundary of the grid forecast data can be more closely fitted to the terrain, thereby solving the above-mentioned problems of jagged edges and difficulty in fitting the boundary.

[0042] In a preferred embodiment, the acquired air pressure value is substituted into the digital elevation of the terrain topology data, and then the terrain topology data after the air pressure value is substituted is interpolated into the grid forecast data in the terrain following coordinate system by a triangulation method, so as to further improve the prediction accuracy. Digital elevation is also called digital elevation model. Digital elevation model is a kind of physical ground model that represents the ground elevation in the form of a set of ordered numerical arrays. It realizes the digital simulation of the ground terrain through limited terrain elevation data. It is generally used to describe the distribution of various geomorphic factors including elevation, such as the spatial distribution of linear and nonlinear combinations of factors such as slope, aspect, slope change rate, etc.

[0043] In addition, in the tile data formed by the terrain topology data, if the area corresponding to the wind farm spans the entire grid points, that is, the area corresponding to the wind farm happens to be distributed on the edges of several adjacent tile data, it is necessary to merge tiles and crop the area of ​​the terrain topology data to obtain terrain topology data that can be used for subsequent interpolation.

[0044] The triangulation method used in this embodiment is a method of dividing the plane corresponding to the data into triangles when the discrete point data is gridded. The gridding of discrete point data generally refers to the role of the triangulation method. Specifically, in the terrain topology data, the data points are denser where the terrain changes faster, and the data points are sparser where the terrain changes slower. In addition, the ground height of each data point in the terrain topology data is not an integer multiple of a grid unit, and most pixel points will not fall exactly on the neat height axis grid. Therefore, the position of each data point in the terrain topology data relative to the neat orthogonal coordinate system grid is a "discrete point".

[0045] For a plane point set P = {P1, P2, ..., P n} plane, the triangle set T obtained by triangulation method is T = {T1, T2, …, T n}, the following conditions must be met:

[0046] (1) The endpoints of all triangles form the set P.

[0047] (2) The sides of any two triangles do not intersect (either they overlap or they have no intersection).

[0048] (3) The union of all triangles constitutes the convex hull of the plane point set P.

[0049] (4) The circumscribed circles of all triangles (excluding the boundaries) do not contain any vertices in the plane point set P.

[0050] When the above conditions are met, the triangulation method is used to make each data point fall independently in a triangle that is not contained in the circumscribed circle, and the value of the interpolation point is obtained by the weighted average of the distances of the data in all the triangular areas surrounding the interpolation point.

[0051] In a specific embodiment, when the terrain topology data is interpolated into the grid forecast data through the triangulation method, the parameters such as soil type, land utilization rate, soil temperature, vegetation number, albedo, etc. in the terrain topology data will also be interpolated into the regular grid of the grid forecast data, thereby improving the accuracy of the disaster prediction results.

[0052] In one embodiment, the step S201 includes:

[0053] The grid forecast data is mapped from the orthogonal coordinate system to the terrain following coordinate system according to the following formula:

[0054] x * =x

[0055] y * =y

[0056] σ=p / p0

[0057] t * =t

[0058] Wherein, x, y, and t represent the horizontal axis coordinate component, vertical axis coordinate component, and time coordinate component of the grid forecast data in the orthogonal coordinate system, respectively. * ,y * ,t * They respectively represent the horizontal axis coordinate component, vertical axis coordinate component and time coordinate component of the grid forecast data in the terrain following coordinate system, σ represents the pressure coordinate component of the grid forecast data in the terrain following coordinate system, p represents the air pressure value of the spatial point, and p0 represents the ground pressure value.

[0059] In this embodiment, the grid forecast data is mapped from the orthogonal coordinate system (x, y, z, t) to the terrain following coordinate system (x * ,y * ,σ,t * ), where z represents the height, the spatial point pressure value p is obtained based on the height of the spatial point, and the pressure coordinate component σ is calculated by combining the spatial point pressure value p and the ground pressure value p0 to complete the coordinate system conversion of the grid forecast data.

[0060] In one embodiment, the step S202 includes:

[0061] Calculate the air pressure value p at different heights of the wind farm according to the following formula: z :

[0062]

[0063] Among them, p ref It represents the ground pressure value, z is the height, R is the air constant, T is the ambient temperature, e is the base of the natural logarithm, and g is the acceleration of gravity.

[0064] In this embodiment, the ground pressure value p of the area corresponding to the wind farm is ref Combined with the corresponding air constant R, ambient temperature T, base of natural logarithm e and gravitational acceleration g, the air pressure value p at different heights in the corresponding area of ​​the wind farm is obtained. z It is understandable that the ground pressure value p in the area corresponding to the wind farm ref , air constant R, ambient temperature T, base of natural logarithm e and gravitational acceleration g can be obtained from the meteorological element data corresponding to the wind farm.

[0065] In other specific embodiments, the gravity acceleration g of the area corresponding to the wind farm may also be calculated according to the following formula:

[0066] g=9.7803(1+0.0053024sin2ψ-0.000005sin22ψ)

[0067] Here, ψ represents the geographical latitude.

[0068] In one embodiment, in combination Figure 3 As shown, the step S103 includes steps S301 to S303.

[0069] S301, performing semantic analysis on the weather forecast data by a text event driven method, extracting event keywords, and constructing event variables according to the event keywords;

[0070] S302, performing numerical forecast integral calculation according to the grid element data set to obtain a regional forecast, and performing statistics on the regional forecast to obtain statistical diagnostic information;

[0071] S303: predicting wind farm disasters in combination with the event variables and statistical diagnosis information to generate the disaster prediction result.

[0072] In this embodiment, firstly, semantic analysis is performed on weather forecast data by a text event-driven method to extract event keywords about the weather conditions of wind farms, and event variables for predicting wind farm disasters are constructed based on the event keywords; secondly, numerical forecast integral calculation is performed based on the grid element data set to obtain regional forecasts for wind farms, and the regional forecasts are statistically analyzed to obtain statistical diagnostic information; finally, the event variables obtained based on the weather forecast data and the statistical diagnostic information obtained based on the grid element data set are combined, and wind farm disasters are predicted based on the combined results to generate disaster prediction results.

[0073] In a specific application scenario, an event variable may refer to a (L, E, P, T) four-variable tuple, which means that a significant weather phenomenon E appears at position L at time T, and variable P is used to store the numerical parameter list of the four-variable tuple. The process of extracting corresponding keywords from sentences and then constructing event variables based on keywords can first use machine learning methods to perform semantic analysis (extract event phrases), analyze sentences related to the weather conditions of wind farms from weather forecast data, such as screening out sentences describing weather conditions such as "wind, rain, snow, and cooling", and then use syntactic analysis methods to find predicates and verbs in sentences, and find sentences that start with verbs, end with prepositions, and have continuous word sequences in the middle. Then, variables are extracted from the sentence, and phrases containing places such as "city, county, district, town" in the sentence are matched with variable L. The longitude and latitude corresponding to the place are found in the dictionary and assigned to variable L. Based on the located predicate, the first noun containing the object of the sentence is found to the right as variable E, and phrases containing numbers and units are assigned to variable P by searching from the preposition to the left. Variable T is set to the release time of weather forecast data by default. If a phrase containing "sometimes" and "days" is found from the preposition to the right, the phrase is judged as a more accurate time, and the default value is replaced by the time. If it is not found from the preposition to the right, it is searched from the beginning of the sentence to before the object. If it still cannot be found, the default value is used.

[0074] In a specific embodiment, the numerical forecast integral calculation is to start from the initial state and perform time integration on the dynamic and thermodynamic equations that govern the atmospheric motion. For example, the grid element data set at 00:00 on the 19th is taken as the starting point and the grid element data set at that time is taken as the initial state, and then the time integration is performed with 12:00 on the 19th as the end point, and the result is used as the regional forecast. Mathematically speaking, the numerical forecast integral calculation is a step-by-step iterative process according to the time step (for example, every 1 hour, every 10 minutes), and this process can be calculated using the Runge-Kutta algorithm; physically speaking, the numerical forecast integral calculation is to solve the equation set corresponding to the grid element data set, for example, solving the equation set shown below:

[0075]

[0076]

[0077]

[0078]

[0079] Among them, u and v represent the gridded wind speed and wind direction along the horizontal axis; w represents the wind speed component in the vertical direction; q represents the water vapor in the atmosphere; θ represents the heat in the atmosphere; S q and S θ denote the flux of water vapor and heat respectively; ρ0 is the air density. In addition, u, v and w represent the dynamics of the atmosphere. and The magnitude of depends on the viscosity and velocity gradient of the atmospheric air mass. It can represent the effect of the viscous force generated by the air velocity in the y-axis (vertical axis) direction on the momentum of the air mass in the data network. It can represent the effect of the viscous force generated by the air velocity in the x-axis (horizontal axis) direction in the data network on the momentum of the air mass; the horizontal line added to the variable outside the square brackets in the formula represents the Reynolds average of the variable, indicating the average state of atmospheric motion, and the variable with apostrophe in the square brackets represents the turbulence term contained in the atmospheric motion, that is, the short-term change in the data grid.

[0080] The numerical forecast integral calculation can obtain the grid element data set after a period of time from the initial time (the event corresponding to the initial state). For example, if the initial time is t0 and the end time is t1, the time interval of the numerical forecast integral calculation can be set as K, that is, the numerical forecast integral calculation will obtain t 0+k ,t 0+2k ...and other output results at each time until t 0+nk is equal to or greater than t1. Here, t 0+k ,t 0+2k ,...,t 0+nk The output results at each time can be expressed as a grid feature dataset at the corresponding time, or as the initial state at the corresponding time.

[0081] In addition, it should be noted that in this embodiment, there is no order of priority for the steps of constructing event variables and obtaining statistical diagnostic information. In practical applications, the order of these two steps can be swapped or performed simultaneously.

[0082] In one embodiment, step S303 includes:

[0083] Extracting diagnostic variables according to the statistical diagnostic information, matching the diagnostic variables with the event variables, and obtaining matching results;

[0084] A disaster diagnosis is performed on the wind farm according to the matching result to obtain a diagnosis result, and the diagnosis result is used as the disaster prediction result.

[0085] In this embodiment, diagnostic variables are first extracted based on statistical diagnostic information, and then the diagnostic variables are matched with event variables. Wind farm disasters are diagnosed based on the matching results, that is, the disasters that will occur in the wind farm are judged, and finally the diagnosis results are used as disaster prediction results for the wind farm.

[0086] In specific application scenarios, diagnostic variables can be extracted from the grid element dataset at the corresponding time. The diagnostic variables may include parameters such as surface temperature, zonal wind speed, radial wind speed, vertical wind speed, water vapor concentration, potential temperature, non-cumulonimbus precipitation, cumulonimbus precipitation, hail equivalent rainfall, Schawat index, and convective effective potential energy.

[0087] On this basis, when the event variables include low temperature, frost and rainfall, it is determined whether the surface temperature in the diagnostic variable is lower than 0°. If the surface temperature is lower than 0° and the surface temperature drops by more than 8°C within 24 hours, or drops by more than 10°C within 48 hours, it is determined that a low temperature frost disaster will occur in the area corresponding to the wind farm, and the corresponding disaster prediction results are generated for feedback. The disaster prediction results can also include information such as the wind farm number and the forecast time of precipitation.

[0088] When the event variables include hail and rainfall, it is determined whether the hail equivalent rainfall in the diagnostic variable is greater than 0, or whether there is rainfall in the wind farm and the vertical wind speed is greater than 5m / s. If so, it is determined that hail or lightning disasters will occur in the corresponding area of ​​the wind farm, and the corresponding disaster prediction results are generated for feedback. Alternatively, it can be determined whether there is rainfall in the wind farm and the vertical wind speed is greater than 5m / s, and at the same time, the Schawat index <0 and the convective effective potential energy> 2000J / kg. If so, it is determined that hail or lightning disasters will occur in the corresponding area of ​​the wind farm.

[0089] When the event variables include strong wind and heavy wind, it is determined whether the horizontal wind speed within 200 meters above the wind farm is greater than the preset warning wind speed. If so, it is determined that a low-altitude strong wind disaster will occur in the corresponding area of ​​the wind farm. 2 =(zonal wind speed) 2 +(Radial wind speed) 2 Calculated.

[0090] In one embodiment, after step S302, the method further includes:

[0091] Numerical results in the regional forecast are extracted to obtain numerical diagnostic information, and the numerical diagnostic information is visualized in the form of layers.

[0092] In this embodiment, numerical results are first extracted from regional forecasts, and then the numerical results are used as numerical diagnostic information, and the numerical diagnostic information is visualized in the form of layers, so as to more intuitively display the predicted wind farm disasters and facilitate wind farm managers to carry out disaster prevention work.

[0093] In specific application scenarios, the following effects can be achieved through layers:

[0094] (1) Display the equipment layout of the wind farm, such as the location of each wind turbine;

[0095] (2) Use continuous heat map animation to describe the forecast results of each time period of the basic meteorological elements corresponding to the wind farm, and set different display colors according to the degree of disaster;

[0096] (3) Display the disaster information in the weather forecast data at the location of the corresponding weather station;

[0097] (4) Display the disaster prediction results at the corresponding location in the wind farm.

[0098] In a specific embodiment, in combination Figure 4 and Figure 5 As shown, a wind farm disaster warning system 400 can be constructed according to the wind farm disaster warning method provided in an embodiment of the present invention, and the system includes a data acquisition module 401, an NWP (numerical weather prediction) module 402, a variable matching module 403 and a disaster warning module 404. Among them, the NWP module 402 further includes a data integration module 501, a prediction research module 502, and a diagnostic information generation module 503.

[0099] The data acquisition module 401 is used to acquire meteorological element data, weather forecast data, grid forecast data and terrain topology data, and input the meteorological element data, grid forecast data and terrain topology data into the NWP module 402, and input the weather forecast data into the variable matching module 403.

[0100] The data integration module 501 is used to receive meteorological element data, grid forecast data and terrain topology data, and perform fusion, interpolation and other processing to obtain a grid element data set, and then input the grid element data set into the prediction research module 502;

[0101] The prediction research module 502 is used to receive the grid element data set, perform numerical prediction integral calculation according to the grid element data set, obtain the regional forecast, and input the regional forecast to the diagnostic information generation module 503;

[0102] The diagnostic information generation module 503 is used to receive regional forecasts, and extract statistical diagnostic information and numerical diagnostic information based on the regional forecasts, and then input the statistical diagnostic information into the variable matching module 403, and input the numerical diagnostic information into the disaster warning module 404.

[0103] The variable matching module 403 is used to receive statistical diagnostic information and weather forecast data, and generate diagnostic variables and event variables based on the statistical diagnostic information and weather forecast data for matching, and diagnose wind farm disasters according to the matching results, and finally send the diagnosis results to the disaster warning module 404.

[0104] The disaster warning module 404 is used to receive numerical diagnosis information and diagnosis results, and perform visual display and disaster warning respectively.

[0105] Figure 6 A schematic block diagram of a wind farm disaster warning device 600 provided in this embodiment, the device 600 includes:

[0106] The data acquisition unit 601 is used to acquire meteorological element data, weather forecast data, grid forecast data and terrain topology data;

[0107] A data integration unit 602 is used to fuse the meteorological element data into the terrain topology data, and interpolate the fused terrain topology data into the grid forecast data to obtain a grid element data set;

[0108] The disaster prediction unit 603 is used to extract events from the weather forecast data to obtain event variables, perform statistics on the grid element data set to obtain diagnostic information, and match the event variables with the diagnostic information to generate disaster prediction results and issue warnings.

[0109] In one embodiment, in combination Figure 7 As shown, the data integration unit 602 includes:

[0110] A coordinate mapping unit 701 is used to map the grid forecast data from an orthogonal coordinate system to a terrain following coordinate system to obtain the grid forecast data in the terrain following coordinate system;

[0111] An air pressure acquisition unit 702, configured to acquire air pressure values ​​at different heights of the wind farm according to the meteorological element data, and substitute the air pressure values ​​into the terrain topology data;

[0112] The data interpolation unit 703 is used to interpolate the terrain topology data after the air pressure value is substituted into the grid forecast data in the terrain following coordinate system through a triangulation method to obtain the grid element data set.

[0113] In one embodiment, the coordinate mapping unit 701 includes:

[0114] The coordinate mapping calculation unit maps the grid forecast data from the orthogonal coordinate system to the terrain following coordinate system according to the following formula:

[0115] x * =x

[0116] y * =y

[0117] σ=p / p0

[0118] t * =t

[0119] Wherein, x, y, and t represent the horizontal axis coordinate component, vertical axis coordinate component, and time coordinate component of the grid forecast data in the orthogonal coordinate system, respectively. * ,y * ,t * They respectively represent the horizontal axis coordinate component, vertical axis coordinate component and time coordinate component of the grid forecast data in the terrain following coordinate system, σ represents the pressure coordinate component of the grid forecast data in the terrain following coordinate system, p represents the air pressure value of the spatial point, and p0 represents the ground pressure value.

[0120] In one embodiment, the air pressure acquisition unit 702 includes:

[0121] The air pressure calculation unit is used to calculate the air pressure value p at different heights of the wind farm according to the following formula z :

[0122]

[0123] Among them, p ref It represents the ground pressure value, z is the height, R is the air constant, T is the ambient temperature, e is the base of the natural logarithm, and g is the acceleration of gravity.

[0124] In one embodiment, in combination Figure 8 As shown, the data interpolation unit 703 includes:

[0125] The variable construction unit 801 is used to perform semantic analysis on the weather forecast data by a text event-driven method, extract event keywords, and construct event variables according to the event keywords;

[0126] An integral calculation unit 802 is used to perform numerical forecast integral calculation according to the grid element data set to obtain a regional forecast, and to perform statistics on the regional forecast to obtain statistical diagnostic information;

[0127] The result generating unit 803 is used to predict the wind farm disaster by combining the event variables and the statistical diagnosis information to generate the disaster prediction result.

[0128] In one embodiment, the result generating unit 803 includes:

[0129] a variable matching unit, configured to extract diagnostic variables according to the statistical diagnostic information, and match the diagnostic variables with the event variables to obtain matching results;

[0130] The disaster diagnosis unit is used to perform disaster diagnosis on the wind farm according to the matching result, obtain a diagnosis result, and use the diagnosis result as the disaster prediction result.

[0131] In one embodiment, the wind farm disaster warning device 600 further includes:

[0132] The visualization unit is used to extract the numerical results in the regional forecast to obtain numerical diagnostic information, and visualize the numerical diagnostic information in the form of layers.

[0133] 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.

[0134] The embodiment of the present invention further provides a computer-readable storage medium on which a computer program is stored, and when the computer program is executed, the steps provided in the above embodiment can be implemented. The storage medium may include: a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and other media that can store program codes.

[0135] The embodiment of the present invention also provides a computer device, which may include a memory and a processor, wherein a computer program is stored in the memory, and when the processor calls the computer program in the memory, the steps provided in the above embodiment may be implemented. Of course, the computer device may also include various network interfaces, power supplies and other components.

[0136] 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 various embodiments can be referred to each other. For the system 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 this 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 this application.

[0137] 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 "comprises", "comprising" 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 "comprising a ..." does not exclude the presence of other identical elements in the process, method, article or device including the element.

Claims

1. A wind farm disaster early warning method, characterized in that: include: Acquire meteorological element data, weather forecast data, grid forecast data and terrain topology data; wherein the weather forecast data includes a weather bulletin for the location of the wind farm; Fusing the meteorological element data into the terrain topology data, and interpolating the fused terrain topology data into the grid forecast data to obtain a grid element data set; Performing event extraction on the weather forecast data to obtain event variables, performing statistics on the grid element data set to obtain diagnostic information, and matching the event variables with the diagnostic information to generate disaster prediction results and issue warnings; The method comprises: performing event extraction on the weather forecast data to obtain event variables, performing statistics on the grid element data set to obtain diagnostic information, and matching the event variables with the diagnostic information to generate disaster prediction results and issue early warnings, including: performing semantic analysis on the weather forecast data through a text event-driven method to extract event keywords, and constructing event variables based on the event keywords; performing numerical forecast integral calculation based on the grid element data set to obtain a regional forecast, and performing statistics on the regional forecast to obtain statistical diagnostic information; and predicting wind farm disasters in combination with the event variables and the statistical diagnostic information to generate the disaster prediction results.

2. The wind farm disaster early warning method according to claim 1, characterized in that: The step of fusing the meteorological element data into the terrain topology data, and interpolating the fused terrain topology data into the grid forecast data to obtain a grid element data set includes: Mapping the grid forecast data from the orthogonal coordinate system to the terrain following coordinate system to obtain the grid forecast data in the terrain following coordinate system; Acquire air pressure values ​​at different heights of the wind farm according to the meteorological element data, and substitute the air pressure values ​​into the terrain topology data; The terrain topology data after the air pressure value is substituted is interpolated into the grid forecast data in the terrain following coordinate system through a triangulation method to obtain the grid element data set.

3. The wind farm disaster early warning method according to claim 2, characterized in that: Mapping the grid forecast data from the orthogonal coordinate system to the terrain following coordinate system to obtain the grid forecast data in the terrain following coordinate system includes: The grid forecast data is mapped from the orthogonal coordinate system to the terrain following coordinate system according to the following formula: x * =x and * =and σ=p / p0 t * =t Wherein, x, y, and t represent the horizontal axis coordinate component, vertical axis coordinate component, and time coordinate component of the grid forecast data in the orthogonal coordinate system, respectively. * ,y * ,t * They respectively represent the horizontal axis coordinate component, vertical axis coordinate component and time coordinate component of the grid forecast data in the terrain following coordinate system, σ represents the pressure coordinate component of the grid forecast data in the terrain following coordinate system, p represents the air pressure value of the spatial point, and p0 represents the ground pressure value.

4. The wind farm disaster early warning method according to claim 2, characterized in that: The step of obtaining the air pressure at different heights of the wind farm according to the meteorological element data includes: Calculate the air pressure value p at different heights of the wind farm according to the following formula: z : Among them, p ref It represents the ground air pressure value, z is the height, R is the air constant, T is the ambient temperature, e is the base of the natural logarithm, and g is the acceleration of gravity.

5. The wind farm disaster early warning method according to claim 1, characterized in that: The predicting of wind farm disasters by combining the event variables and the statistical diagnostic information to generate the disaster prediction results includes: Extracting diagnostic variables according to the statistical diagnostic information, matching the diagnostic variables with the event variables, and obtaining matching results; A disaster diagnosis is performed on the wind farm according to the matching result to obtain a diagnosis result, and the diagnosis result is used as the disaster prediction result.

6. The wind farm disaster early warning method according to claim 1, characterized in that: After obtaining the regional forecast, the method further includes: Numerical results in the regional forecast are extracted to obtain numerical diagnostic information, and the numerical diagnostic information is visualized in the form of layers.

7. A wind farm disaster warning device, characterized in that: include: A data acquisition unit, used to acquire meteorological element data, weather forecast data, grid forecast data and terrain topology data; wherein the weather forecast data includes a weather bulletin for the location of the wind farm; A data integration unit, used for fusing the meteorological element data into the terrain topology data, and interpolating the fused terrain topology data into the grid forecast data to obtain a grid element data set; A disaster prediction unit, configured to extract events from the weather forecast data to obtain event variables, perform statistics on the grid element data set to obtain diagnostic information, match the event variables with the diagnostic information, generate disaster prediction results and issue warnings; The disaster prediction unit includes: a variable construction unit, which is used to perform semantic analysis on the weather forecast data through a text event-driven method, extract event keywords, and construct event variables based on the event keywords; an integral calculation unit, which is used to perform numerical forecast integral calculation based on the grid element data set to obtain a regional forecast, and perform statistics on the regional forecast to obtain statistical diagnostic information; and a result generation unit, which is used to predict wind farm disasters in combination with the event variables and statistical diagnostic information to generate the disaster prediction results.

8. A computer device, characterized in that: The method comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the wind farm disaster warning method according to any one of claims 1 to 6 when executing the computer program.

9. 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 wind farm disaster early warning method according to any one of claims 1 to 6 is implemented.

Images