Wind speed mutation early warning method and device for wind farm and computer equipment

Abstract

The application provides a wind farm wind speed mutation early warning method and device and computer equipment, and relates to the wind power generation technical field. The application detects whether the wind speed mutation wind condition occurs at the current wind farm wind speed mutation early warning model based on the actual wind turbine operation data of each wind turbine in the target wind farm, and extracts the target turbine identifier and the target early warning time length of the target turbine identifier at the wind turbine from the target wind farm wind speed mutation early warning model when the wind speed mutation wind condition is detected. Then, the target wind turbine pointed to by the target turbine identifier is warned to encounter the wind speed mutation wind condition at the future time interval from the current time by the target early warning time length, so that the spatial correlation relationship between different wind turbines in the target wind farm is utilized to early warn the wind speed mutation of each wind turbine.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of wind power generation, in particular to a wind farm wind speed mutation early warning method and device and computer equipment. BACKGROUND

[0002] With the continuous development of science and technology, wind power generation technology is widely used in mountainous, offshore and other wind farms. For mountainous wind farms, the wind conditions of mountainous wind farms often change frequently due to geographical environment and terrain factors, resulting in that the operating conditions of wind turbines in the mountainous wind farms cannot match the actual wind conditions. For example, when the wind speed of the windward surface of the wind turbine in the mountainous wind farm suddenly rises, the corresponding turbine load will rise sharply, resulting in a decrease in the fatigue life of the corresponding turbine components. When the wind speed of the windward surface of the wind turbine in the mountainous wind farm suddenly drops, the corresponding wind turbine fails to pitch in time, resulting in a loss of power generation of the corresponding wind turbine. Therefore, in order to effectively ensure that the operating conditions of the wind turbines in the mountainous wind farm match the actual wind conditions of the mountainous wind farm, it is particularly important to give early warning of the wind speed mutation of the entire mountainous wind farm. SUMMARY

[0003] Therefore, the purpose of the present application is to provide a wind farm wind speed mutation early warning method and device and computer equipment, which can give early warning of wind speed mutation for each wind turbine based on the spatial correlation relationship between different wind turbines in the target wind farm, so as to control the operating conditions of each wind turbine in the entire target wind farm to a state substantially matching the actual wind conditions of the target wind farm, and improve the operating stability and power generation efficiency of each wind turbine in the target wind farm.

[0004] In order to achieve the above purpose, the technical solutions adopted by the embodiments of the present application are as follows:

[0005] In a first aspect, the present application provides a wind farm wind speed mutation early warning method, which comprises:

[0006] obtaining actual turbine operating data of all wind turbines in a target wind farm at a current time;

[0007] for each wind turbine in the target wind farm, based on the actual turbine operating data of the wind turbine, calling a target wind speed mutation early warning model matched with the target wind farm to detect whether a wind speed mutation wind condition occurs at the wind turbine at present;

[0008] in the case where the target wind speed mutation early warning model detects that a wind speed mutation wind condition occurs at the wind turbine at present, extracting a target turbine identifier with the strongest spatial correlation with the wind turbine and a target early warning time length of the target turbine identifier at the wind turbine from the target wind speed mutation early warning model.

[0009] sending a wind speed sudden change early warning message to a target wind turbine identified by the target group identifier, wherein the wind speed sudden change early warning message is used to warn the target wind turbine that a wind speed sudden change weather condition will occur at a future time interval of the target early warning time length from a current time.

[0010] In an optional embodiment, the step of detecting whether a wind speed sudden change weather condition occurs at the wind turbine based on the actual wind turbine operation data of the wind turbine and the target wind speed sudden change early warning model matched with the target wind farm, comprises:

[0011] calculating a current wind speed sudden change feature of the wind turbine based on a wind speed sudden change feature calculation strategy recorded by the target wind speed sudden change early warning model;

[0012] calculating a sudden change feature difference between the current wind speed sudden change feature of the wind turbine and a plurality of historical wind speed sudden change features of the wind turbine in a historical time period close to a current time;

[0013] detecting whether the plurality of sudden change feature differences satisfy a plurality of wind speed sudden change boundary conditions;

[0014] in a case where it is detected that the plurality of sudden change feature differences satisfy at least one wind speed sudden change boundary condition, determining that a wind speed sudden change weather condition occurs at the wind turbine.

[0015] In an optional embodiment, the method further comprises:

[0016] obtaining a plurality of wind speed sudden change sample sets respectively including a plurality of wind turbines in different wind farm sectors of the target wind farm in a preset historical time period, wherein each wind speed sudden change sample set includes wind speed sudden change samples corresponding to the corresponding wind turbine at different historical wind speed sudden change times in the preset historical time period, each wind speed sudden change sample includes historical wind turbine operation data of the corresponding wind turbine near the corresponding historical wind speed sudden change time, and a historical wind speed sudden change feature of the corresponding wind turbine calculated based on a wind speed sudden change feature calculation strategy at the corresponding historical wind speed sudden change time;

[0017] for each wind turbine in the target wind farm, performing a sliding spatial correlation analysis on the wind turbine and other wind turbines belonging to the same wind farm sector as the wind turbine according to the wind speed sudden change sample set of the wind turbine, to obtain a target group identifier of a target wind turbine in the corresponding wind farm sector having the strongest spatial correlation with the wind turbine, and a target early warning time length and an actual spatial correlation coefficient of the target group identifier at the wind turbine;

[0018] According to the wind speed mutation feature calculation strategy and historical unit operation data of the wind turbine in the preset historical time period, a unit wind speed mutation early warning model matched with the target unit identifier of the wind turbine is constructed based on the target unit identifier, the target early warning time length and the actual spatial correlation coefficient corresponding to the wind turbine, wherein the unit wind speed mutation early warning model is used for wind speed mutation early warning of a target wind turbine pointed to by the target unit identifier.

[0019] The model integration processing is performed on the unit wind speed mutation early warning models corresponding to all wind turbines of the target wind farm respectively, to obtain a target wind speed mutation early warning model matched with the target wind farm.

[0020] In an optional implementation, the step of obtaining wind speed mutation sample sets of a plurality of wind turbines included in different wind farm sectors of the target wind farm respectively in a preset historical time period, comprises:

[0021] For each wind turbine in the target wind farm, historical unit operation data of the wind turbine at all historical time points in the preset historical time period are obtained;

[0022] Based on the wind speed mutation feature calculation strategy, historical wind speed mutation features corresponding to the historical unit operation data of the wind turbine at different historical time points are calculated;

[0023] For each historical time point in the preset historical time period, it is detected whether the historical wind speed mutation feature of the wind turbine at the corresponding historical time point appears wind speed mutation weather condition;

[0024] In a case where it is detected that the historical wind speed mutation feature of the wind turbine at the corresponding historical time point appears wind speed mutation weather condition, the historical time point is taken as a historical wind speed mutation time point of the wind turbine, and historical unit operation data of the wind turbine near the historical wind speed mutation time point and the historical wind speed mutation feature of the wind turbine at the historical wind speed mutation time point are integrated into a wind speed mutation sample of the wind turbine.

[0025] In an optional implementation, the step of detecting whether the historical wind speed mutation feature of the wind turbine at the corresponding historical time point appears wind speed mutation weather condition, comprises:

[0026] The mutation feature difference between the historical wind speed mutation feature of the wind turbine at the corresponding historical time point and a plurality of historical wind speed mutation features of the wind turbine in a historical time period near the corresponding historical time point is calculated;

[0027] It is detected whether the calculated plurality of mutation feature differences satisfy a plurality of pre-stored wind speed mutation boundary conditions;

[0028] In the case that the plurality of mutation feature difference values satisfy at least one wind speed mutation boundary condition, it is determined that the wind turbine is in a wind speed mutation weather condition at the historical wind speed mutation feature corresponding to the historical moment.

[0029] In an optional embodiment, for each wind turbine in the target wind farm, the step of performing a sliding spatial correlation analysis on the wind speed mutation sample set of the wind turbine and other wind turbines belonging to the same wind farm sector as the wind turbine, obtaining the target turbine identifier with the strongest spatial correlation with the wind turbine in the corresponding wind farm sector, and the target early warning time length and actual spatial correlation coefficient of the target turbine identifier at the wind turbine, comprises:

[0030] For each wind speed mutation sample in the wind speed mutation sample set of the wind turbine, performing a sliding spatial correlation analysis on the historical turbine operation data of the wind speed mutation sample and other wind turbines belonging to the same wind farm sector as the wind turbine in the preset historical time period, obtaining a spatial correlation matrix of the wind turbine at the wind speed mutation sample, wherein the spatial correlation matrix comprises the spearman correlation coefficients between the corresponding wind speed mutation sample of the wind turbine and each other wind turbine at different sliding window time steps;

[0031] The largest spearman correlation coefficient in the spatial correlation matrix of the wind turbine at the wind speed mutation sample is taken as the optimal spatial correlation coefficient of the wind turbine at the wind speed mutation sample, and the sliding window time step corresponding to the optimal spatial correlation coefficient and the turbine identifier of the other wind turbine are taken as the optimal early warning time length and the optimal strong correlation turbine identifier of the wind turbine at the wind speed mutation sample, respectively.

[0032] Based on the principle of majority over minority, the optimal strong correlation turbine identifier, the optimal early warning time length and the optimal spatial correlation coefficient corresponding to each wind speed mutation sample of the wind turbine are processed to obtain the target turbine identifier with the strongest spatial correlation with the wind turbine, and the target early warning time length and the actual spatial correlation coefficient of the target turbine identifier at the wind turbine.

[0033] In an optional embodiment, the wind speed mutation feature calculation strategy is represented by a function expression satisfying function continuity, and the function expression is as follows:

[0034] f(Ω i ,V t ,β t ,W t )=ω1*V t +ω2*β t +ω3*W t;

[0035] wherein f(Ω i ,V t ,β t ,W t ) is used to represent the wind speed mutation feature of the i-th wind turbine at the t-th moment, Ω i is used to represent the nonlinear gain vector of the i-th wind turbine adapted to the wind power curve, V t is used to represent the wind speed included in the unit operation data of the i-th wind turbine at the t-th moment, β t is used to represent the pitch angle included in the unit operation data of the i-th wind turbine at the t-th moment, W t is used to represent the power included in the unit operation data of the i-th wind turbine at the t-th moment, wherein Ω i =[ω1,ω2,ω3], ω1, ω2 and ω3 are respectively used to represent the actual nonlinear gain size of the i-th wind turbine adapted to the wind power curve.

[0036] In a second aspect, the present application provides a wind speed mutation early warning device of a wind farm, the device comprising:

[0037] an operation parameter acquisition module, configured to acquire actual unit operation data of all wind turbines in a target wind farm at a current moment;

[0038] a wind speed mutation detection module, configured to, for each wind turbine in the target wind farm, based on the actual unit operation data of the wind turbine, call a target wind speed mutation early warning model matched with the target wind farm to detect whether a wind speed mutation weather condition occurs at the wind turbine at the current moment;

[0039] an early warning parameter extraction module, configured to, in a case where the target wind speed mutation early warning model detects that a wind speed mutation weather condition occurs at the wind turbine at the current moment, extract, from the target wind speed mutation early warning model, a target unit identifier with the strongest spatial correlation with the wind turbine, and a target early warning time length of the target unit identifier at the wind turbine;

[0040] a mutation early warning module, configured to send a wind speed mutation early warning message to a target wind turbine pointed to by the target unit identifier in the target wind farm, wherein the wind speed mutation early warning message is used to alert the target wind turbine to encounter a wind speed mutation weather condition at a future moment spaced apart from the current moment by the target early warning time length.

[0041] In an optional implementation, the device further comprises:

[0042] a mutation sample acquisition module, configured to acquire wind speed mutation sample sets of a plurality of wind turbines respectively included in different wind field sectors of the target wind farm in a preset historical time period, wherein each wind speed mutation sample set includes wind speed mutation samples corresponding to different historical wind speed mutation moments of a corresponding wind turbine in the preset historical time period, each wind speed mutation sample includes historical wind turbine operation data of the corresponding wind turbine near the corresponding historical wind speed mutation moment, and historical wind speed mutation features of the corresponding wind turbine at the corresponding historical wind speed mutation moment calculated based on a wind speed mutation feature calculation strategy;

[0043] a spatial correlation analysis module, configured to, for each wind turbine in the target wind farm, perform a sliding spatial correlation analysis on the wind turbine and other wind turbines belonging to a same wind field sector as the wind turbine according to wind speed mutation sample sets of the wind turbine, to obtain a target turbine identifier of the wind turbine in a corresponding wind field sector having the strongest spatial correlation with the wind turbine, a target early warning time length of the target turbine identifier at the wind turbine, and an actual spatial correlation coefficient;

[0044] a warning model construction module, configured to, according to the wind speed mutation feature calculation strategy and historical wind turbine operation data of the wind turbine in the preset historical time period, construct a wind speed mutation warning model of the wind turbine matching the target turbine identifier based on the target turbine identifier, the target early warning time length, and the actual spatial correlation coefficient of the wind turbine, wherein the wind speed mutation warning model is used for performing wind speed mutation warning on a target wind turbine pointed to by the target turbine identifier;

[0045] a warning model integration module, configured to perform model integration processing on wind speed mutation warning models respectively corresponding to all wind turbines of the target wind farm, to obtain a target wind speed mutation warning model matching the target wind farm.

[0046] In a third aspect, the present application provides a computer device, including a processor and a memory, the memory stores a computer program capable of being executed by the processor, and the processor can execute the computer program to implement the wind farm wind speed mutation warning method in any one of the preceding embodiments.

[0047] In this case, the beneficial effects of the embodiments of the present application can include the following:

[0048] After obtaining actual unit operation data of all wind turbines in the target wind farm at the current time, the application will call the target wind speed mutation early warning model matched with the target wind farm for each wind turbine based on the actual unit operation data of the wind turbine to detect whether the wind speed mutation wind condition occurs at the current time at the wind turbine, and extract the target unit identifier with the strongest spatial correlation of the wind turbine and the target early warning time length of the target unit identifier at the wind turbine from the target wind speed mutation early warning model when detecting that the wind speed mutation wind condition occurs at the current time at the wind turbine. Then, a wind speed mutation early warning message for warning the target wind turbine of the target unit identifier to encounter the wind speed mutation wind condition at a future time interval of the target early warning time length from the current time is sent to the target wind turbine, so as to utilize the spatial correlation relationship between different wind turbines in the target wind farm to perform wind speed mutation early warning for each wind turbine, and control the unit operation condition of each wind turbine in the target wind farm to a state substantially matched with the actual wind condition of the target wind farm, so as to improve the operation stability and power generation efficiency of each wind turbine in the target wind farm.

[0049] In order to make the above objectives, characteristics and advantages of the present application more apparent, the following will describe a preferred embodiment in detail, and the accompanying drawings are referred to as follows. BRIEF DESCRIPTION OF DRAWINGS

[0050] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments. It should be understood that the following drawings only show some embodiments of the present application, and therefore should not be regarded as a limitation to the scope. For those skilled in the art, other related drawings can also be obtained without creative labor on the basis of these drawings.

[0051] Figure 1 The composition schematic diagram of the computer device provided by the embodiments of the present application is shown in the following figure.

[0052] Figure 2 The flowchart of the wind farm wind speed mutation early warning method provided by the embodiments of the present application is shown in the following figure.

[0053] Figure 3 The flowchart of the wind farm wind speed mutation early warning method provided by the embodiments of the present application is shown in the following figure.

[0054] Figure 4 The composition schematic diagram of the wind farm wind speed mutation early warning device provided by the embodiments of the present application is shown in the following figure.

[0055] Figure 5 The composition schematic diagram of the wind farm wind speed mutation early warning device provided by the embodiments of the present application is shown in the following figure.

[0056] Icon: 10-computer device; 11-memory; 12-processor; 13-communication unit; 100-wind farm wind speed sudden change early warning device; 110-operation parameter acquisition module; 120-wind speed sudden change detection module; 130-early warning parameter extraction module; 140-sudden change early warning module; 150-sudden change sample acquisition module; 160-space correlation analysis module; 170-early warning model construction module; 180-early warning model integration module. DETAILED DESCRIPTION

[0057] In order to make the objectives, technical solutions, and advantages of the embodiments of the present application clearer, the following will be a clear and complete description of the technical solutions in the embodiments of the present application in conjunction with the accompanying drawings of the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, rather than all the embodiments. The components of the embodiments of the present application described and shown in the accompanying drawings can be arranged and designed in various different configurations.

[0058] Therefore, the following detailed description of the embodiments of the present application provided in the accompanying drawings is not intended to limit the scope of the claimed present application, but only represents selected embodiments of the present application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present application without creative labor are within the scope of protection of the present application.

[0059] It should be noted that: similar reference numbers and letters represent similar items in the following drawings, therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings.

[0060] In the description of the present application, it should be understood that the relationship terms such as "first" and "second" and the like 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 the entities or operations. Moreover, the terms "include", "contain" or any other variants thereof are intended to cover non-exclusive inclusion, so that the process, method, article or device including a series of elements not only includes those elements, but also includes other elements not explicitly listed or inherent to such process, method, article or device. Without more limitations, the element defined by the statement "including a" does not exclude the presence of another identical element in the process, method, article or device including the element. The specific meaning of the above terms in the present application can be understood by the person of ordinary skill in the art.

[0061] The following will be a detailed description of some embodiments of the present application in conjunction with the accompanying drawings. The following embodiments and features in the embodiments can be combined with each other without conflict.

[0062] Please refer to Figure 1 , Figure 1 is a schematic diagram of the computer device 10 provided in the embodiments of the present application. In the embodiments of the present application, the computer device 10 can be communicatively connected with all wind turbines in a mountain wind farm, so as to obtain the turbine operation data of each wind turbine at different operating time points, and effectively detect whether each wind turbine encounters wind speed mutation wind condition according to the obtained turbine operation data of each wind turbine. Then, based on the spatial correlation relationship between different wind turbines in the mountain wind farm, the wind speed mutation early warning is performed for each wind turbine, so that the wind farm management personnel can adjust the turbine operation condition of each wind turbine in the mountain wind farm to a state substantially matching the actual wind condition of the mountain wind farm, thereby effectively improving the operation stability and power generation efficiency of each wind turbine in the mountain wind farm. The turbine operation data includes the turbine speed, wind speed, propeller angle and power generation power of the corresponding wind turbine at a certain operating time point, etc. The computer device 10 can be, but is not limited to, a tablet computer, a notebook computer, a personal computer and a server, etc.

[0063] In the embodiments of the present application, the computer device 10 can include a memory 11, a processor 12, a communication unit 13 and a wind farm wind speed mutation early warning apparatus 100. The memory 11, the processor 12 and the communication unit 13 are directly or indirectly electrically connected with each other to realize data transmission or interaction. For example, the memory 11, the processor 12 and the communication unit 13 can be electrically connected with each other through one or more communication buses or signal lines.

[0064] In the embodiments, the memory 11 can be, but is not limited to, a random access memory (RAM), a read only memory (ROM), a programmable read only memory (PROM), an erasable programmable read only memory (EPROM), an electrically erasable programmable read only memory (EEPROM) and the like. The memory 11 is used to store a computer program, and the processor 12 can execute the computer program accordingly after receiving an execution instruction.

[0065] In this embodiment, the processor 12 can be an integrated circuit chip with signal processing capabilities. The processor 12 can be a general-purpose processor, including at least one of a Central Processing Unit (CPU), Graphics Processing Unit (GPU), Network Processor (NP), Digital Signal Processor (DSP), Application-Specific Integrated Circuit (ASIC), Field-Programmable Gate Array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. The general-purpose processor can be a microprocessor or any conventional processor, capable of implementing or executing the methods, steps, and logic block diagrams disclosed in the embodiments of this application.

[0066] In this embodiment, the communication unit 13 is used to establish a communication connection between the computer device 10 and other electronic devices through a network, and to send and receive data through the network, wherein the network includes wired communication networks and wireless communication networks. For example, the computer device 10 can use the communication unit 13 to obtain the unit operation data of a wind turbine at different operating times from a wind turbine, or send a wind speed change warning message to another wind turbine located in the same mountain wind farm as the wind turbine.

[0067] In this embodiment, the wind farm wind speed sudden change early warning device 100 includes at least one software function module that can be stored in the memory 11 in the form of software or firmware or in the operating system of the computer device 10. The processor 12 can be used to execute the executable modules stored in the memory 11, such as the software function modules and computer programs included in the wind farm wind speed sudden change early warning device 100. The computer device 10 can use the wind farm wind speed sudden change early warning device 100 to provide early warning of wind speed sudden changes for each wind turbine based on the spatial correlation between different wind turbines in the target wind farm, so as to control the operating status of each wind turbine in the entire target wind farm to a state that substantially matches the actual wind conditions of the target wind farm, thereby improving the operating stability and power generation efficiency of each wind turbine in the target wind farm. The target wind farm is the mountain wind farm to which multiple wind turbines communicate with the computer device 10.

[0068] Understandable, Figure 1 The block diagram shown is only a schematic diagram of one configuration of the computer device 10. The computer device 10 may also include components such as... Figure 1 The more or fewer components shown, or having the same Figure 1 The different configurations shown. Figure 1The components shown in the figures can be implemented in hardware, software, or a combination thereof.

[0069] In the present application, in order to ensure that the computer device 10 can give early warning of wind speed mutation for each wind turbine based on the spatial correlation relationship between different wind turbines in the mountain wind farm, so as to control the operation state of each wind turbine in the entire mountain wind farm to a state substantially matching the actual wind condition of the mountain wind farm, and improve the operation stability and power generation efficiency of each wind turbine in the mountain wind farm, the present embodiment provides a wind farm wind speed mutation early warning method to achieve the foregoing purpose. The wind farm wind speed mutation early warning method provided by the present application will be described in detail below.

[0070] Please refer to Figure 2 , Figure 2 is one of the flowcharts of the wind farm wind speed mutation early warning method provided by the present embodiment. In the present embodiment, the wind farm wind speed mutation early warning method can include steps S210-S240.

[0071] Step S210, obtaining actual unit operation data of all wind turbines in the target wind farm at the current time.

[0072] In the present embodiment, the computer device 10 can send a unit operation data acquisition request to each wind turbine in the target wind farm at a preset time interval to obtain the actual unit operation data of each wind turbine at the current time. The preset time interval can be, but is not limited to, 30S, 1min, or 1.5min, etc.

[0073] Step S220, for each wind turbine in the target wind farm, based on the actual unit operation data of the wind turbine, calling a target wind speed mutation early warning model matched with the target wind farm to detect whether a wind speed mutation wind condition occurs at the wind turbine at present.

[0074] In the present embodiment, the target wind speed mutation early warning model matched with the target wind farm is used to describe the spatial correlation condition of different wind turbines in the target wind farm in the wind speed mutation process, and the time sequence correlation condition of different wind turbines affected by the mutation wind speed in the wind speed mutation process. When the computer device 10 obtains the actual unit operation data of all wind turbines in the target wind farm at the current time, it will input the actual unit operation data of the wind turbine into the target wind speed mutation early warning model based on the characteristics that the wind speed mutation will directly reflect in the unit operation data such as power generation, speed, and pitch angle of the wind turbine, so as to utilize the wind speed mutation early warning model to analyze the input actual unit operation data, and detect whether a wind speed mutation wind condition occurs at the wind turbine at present.

[0075] Optionally, in this embodiment, the target wind speed change early warning model may record a wind speed change feature calculation strategy, which is used to calculate the details of wind speed changes carried by the corresponding turbine operating data. Therefore, for each wind turbine in the target wind farm, the step S220, "based on the actual turbine operating data, calling the target wind speed change early warning model matching the target wind farm to detect whether a wind speed change has occurred at the wind turbine," may include:

[0076] Based on the wind speed change feature calculation strategy recorded by the target wind speed change early warning model, the current wind speed change feature of the wind turbine corresponding to the actual unit operation data is calculated.

[0077] Calculate the difference in abrupt changes in wind speed between the current wind turbine and multiple historical wind speed abrupt changes in wind turbine within a historical time period close to the current moment.

[0078] The system checks whether the calculated differences in multiple abrupt change features satisfy the pre-stored boundary conditions for various wind speed abrupt changes.

[0079] If multiple abrupt change feature differences are detected and satisfy at least one wind speed abrupt change boundary condition, it is determined that a wind speed abrupt change is currently occurring at the wind turbine location.

[0080] In this process, the wind speed abrupt change characteristic calculation strategy is represented by a function expression that satisfies the function continuity, as shown below:

[0081] f(Ω i V t ,β t W t )=ω1*V t +ω2*β t +ω3*W t ;

[0082] Where, f(Ω) i V t ,β t W t ) is used to represent the wind speed change characteristics of the i-th wind turbine at time t, Ω i V is used to represent the nonlinear gain vector of the i-th wind turbine that adapts to the wind power curve. t β is used to represent the wind speed magnitude included in the unit operation data of the i-th wind turbine at time t. t W is used to represent the blade angle size included in the unit operation data of the i-th wind turbine at time t. tsignal i = [ω1, ω2, ω3], ω1, ω2 and ω3 are respectively used to represent the actual nonlinear gain size of the i-th wind turbine adapted to the wind power curve.

[0083] The multiple historical wind speed mutation characteristics of the wind turbine in the historical time period close to the current time can include the historical wind speed mutation characteristics corresponding to the previous one, two, three, five historical time points of the current time of the wind turbine, and the multiple wind speed mutation boundary conditions are used to describe the difference distribution of the wind speed mutation characteristics when the wind speed mutation phenomenon acts on the multiple wind speed mutation characteristics.

[0084] In an embodiment of the embodiment, the multiple wind speed mutation boundary conditions can include the following contents:

[0085] (1) signal t -signal t-3 > θ1, wherein θ1> 0;

[0086] (2) signal t -signal t-5 > θ2, wherein θ2> 0;

[0087]

[0088] (4) signal t -signal t-3 < θ4, wherein θ4< 0;

[0089] (5) signal t -signal t-3 < θ5, wherein θ5< 0;

[0090]

[0091] wherein, signal t is used to represent the wind speed mutation characteristic of the corresponding wind turbine at the t-th time point, signal t-1 is used to represent the wind speed mutation characteristic of the corresponding wind turbine at the t-1-th time point, signal t-2 is used to represent the wind speed mutation characteristic of the corresponding wind turbine at the t-2-th time point, signal t-3 is used to represent the wind speed mutation characteristic of the corresponding wind turbine at the t-3-th time point, and signal t-5is used to represent the wind speed mutation characteristics of the corresponding wind turbine at the t-5th moment. If the difference between the above-mentioned five wind speed mutation characteristics satisfies at least one of the above-mentioned six wind speed mutation boundary conditions, it indicates that the corresponding wind turbine appears the wind speed mutation wind condition at the tth moment.

[0092] Therefore, the present application can effectively detect whether each wind turbine in the target wind farm appears the wind speed mutation wind condition at the current moment by executing the specific step flow of the above-mentioned step S220.

[0093] Step S230, in the case that the target wind speed mutation early warning model is called to detect that the wind speed mutation wind condition currently appears at the wind turbine, the target turbine identifier with the strongest spatial correlation with the wind turbine and the target early warning time length of the target turbine identifier at the wind turbine are extracted from the target wind speed mutation early warning model.

[0094] In the embodiment, if the computer device 10 determines that a certain wind turbine a in the target wind farm currently appears the wind speed mutation wind condition, it indicates that another wind turbine b with the strongest spatial correlation with the wind turbine a in the spatial distribution dimension will also be affected by the wind speed mutation wind condition at the wind turbine a. At this time, the computer device 10 extracts the target turbine identifier (i.e. the turbine ID of the target wind turbine b) with the strongest spatial correlation with the wind turbine a and the target early warning time length of the target turbine identifier at the wind turbine a from the target wind speed mutation early warning model, wherein the target early warning time length is used to represent the delay time length of the target wind turbine b affected by the wind speed mutation wind condition at the aforementioned wind turbine a.

[0095] Step S240, a wind speed mutation early warning message is sent to the target wind turbine pointed to by the target turbine identifier in the target wind farm, wherein the wind speed mutation early warning message is used to alert the target wind turbine to encounter the wind speed mutation wind condition at the future moment separated from the current moment by the target early warning time length.

[0096] In the embodiment, when the computer device 10 determines the target turbine identifier of the target wind turbine b with the strongest spatial correlation with the wind turbine a currently appearing the wind speed mutation wind condition in the target wind farm and the target early warning time length of the target wind turbine b relative to the wind turbine a, a wind speed mutation early warning message is generated for the target wind turbine b based on the current moment, and the wind speed mutation early warning message is sent to the target wind turbine b to alert the target wind turbine b to encounter the wind speed mutation wind condition at the future moment separated from the current moment by the target early warning time length.

[0097] Therefore, the application can perform the steps S210-S240 to give early warning of wind speed mutation for each wind turbine based on the spatial correlation relationship between different wind turbines in the mountain wind farm, so as to control the operation state of each wind turbine in the mountain wind farm to substantially match the actual wind condition of the mountain wind farm, and improve the operation stability and power generation efficiency of each wind turbine in the mountain wind farm.

[0098] Optionally, referring to Figure 3 , Figure 3 is a flowchart of the wind speed mutation early warning method provided by the embodiment of the application. Compared with the wind speed mutation early warning method shown in Figure 2 , the wind speed mutation early warning method shown in Figure 3 may further include steps S250-S280 to further dig the spatial correlation conditions between the multiple wind turbines in different wind field sectors of the target wind farm and the time sequence correlation conditions of the multiple wind turbines in the same wind field sector affected by the same wind speed mutation condition, and construct a target wind speed mutation early warning model for ensuring that the early warning effect of the wind speed mutation of the target wind farm reaches the best state based on the dug spatial correlation conditions and time sequence correlation conditions.

[0099] In step S250, wind speed mutation sample sets of multiple wind turbines included in different wind field sectors of the target wind farm in a preset historical time period are obtained.

[0100] In the embodiment, the target wind farm can be divided into multiple wind field sectors according to different wind directions, and multiple wind turbines are deployed in each wind field sector, and each wind turbine corresponds to a wind speed mutation sample set. Each wind speed mutation sample set can include wind speed mutation samples corresponding to different historical wind speed mutation moments (for example, t0, …, t n ) of the corresponding wind turbine in the preset historical time period, and each wind speed mutation sample includes historical unit operation data of the corresponding wind turbine near the corresponding historical wind speed mutation moment (for example, t0), for example, 10 minutes before and after t0, and historical wind speed mutation features of the corresponding wind turbine calculated based on a wind speed mutation feature calculation strategy at the corresponding historical wind speed mutation moment (for example, t0).

[0101] Optionally, in an implementation of the embodiment, the step S240 can include:

[0102] For each wind turbine in the target wind farm, historical unit operation data of the wind turbine at all historical moments in the preset historical time period is obtained.

[0103] calculating, based on the wind speed mutation feature calculation strategy, the historical wind speed mutation features of the wind turbine at different historical time points respectively corresponding to the historical unit operation data;

[0104] For each historical time point in the preset historical time period, detecting whether the historical wind speed mutation feature of the wind turbine at the corresponding historical time point appears a wind speed mutation weather condition;

[0105] In the case where it is detected that the historical wind speed mutation feature of the wind turbine at the corresponding historical time point appears a wind speed mutation weather condition, taking the historical time point as a historical wind speed mutation time point of the wind turbine, and setting the historical unit operation data of the wind turbine near the historical wind speed mutation time point and the historical wind speed mutation feature of the wind turbine at the historical wind speed mutation time point as a wind speed mutation sample of the wind turbine.

[0106] Wherein, after calculating the historical wind speed mutation features of a certain wind turbine c at different historical time points respectively corresponding thereto, for the historical time points respectively involved in all the calculated historical wind speed mutation features, it can be referred to the specific step process of step S220 to detect whether the historical wind speed mutation feature of the wind turbine c at each historical time point respectively corresponding thereto appears a wind speed mutation weather condition, at this time, for each historical time point in the preset historical time period, the step of detecting whether the historical wind speed mutation feature of the wind turbine c at the corresponding historical time point appears a wind speed mutation weather condition can include:

[0107] calculating the mutation feature difference between the historical wind speed mutation feature of the wind turbine at the corresponding historical time point and the historical wind speed mutation features of the wind turbine in the historical time period near the corresponding historical time point;

[0108] detecting whether the calculated multiple mutation feature differences satisfy the pre-stored multiple wind speed mutation boundary conditions;

[0109] In the case where it is detected that the multiple mutation feature differences satisfy at least one wind speed mutation boundary condition, it is determined that the historical wind speed mutation feature of the wind turbine at the corresponding historical time point appears a wind speed mutation weather condition.

[0110] Taking historical time point t0 as an example, the historical wind speed mutation feature of the wind turbine c at the corresponding historical time point t0 is signal t0 , and the multiple historical wind speed mutation features of the wind turbine c in the historical time period near the corresponding historical time point t0 can include the historical wind speed mutation features respectively corresponding to the first, second, third and fifth historical time points before the historical time point t0 of the wind turbine c (i.e. signal t0-1 , signal t0-2 , signal t0-3 , and signal t0-5If the difference between the five historical wind speed mutation characteristics satisfies at least one of the six wind speed mutation boundary conditions, it indicates that the wind turbine c has a wind speed mutation weather condition at the historical wind speed mutation characteristic corresponding to the historical time t0, and the historical time t0 is a historical wind speed mutation time of the wind turbine c in the preset time period.

[0111] Therefore, the present application can collect the wind speed mutation sample set of the wind speed mutation weather condition of the wind turbine in the preset historical time period in the target wind farm through the specific step flow of step S250, so as to directly use the wind speed mutation sample set related to the wind speed mutation weather condition for subsequent sliding space correlation analysis, thereby improving the actual analysis efficiency and analysis result accuracy of the subsequent sliding space correlation analysis operation, avoiding the influence of a large amount of historical unit operation data irrelevant to the wind speed mutation weather condition on the final overall analysis time consumption and analysis result accuracy in the sliding space correlation analysis process, and effectively improving the model training efficiency and early warning result accuracy of the target wind speed mutation early warning model.

[0112] Step S260, for each wind turbine in the target wind farm, according to the wind speed mutation sample set of the wind turbine, the sliding space correlation analysis is performed on the wind turbine and other wind turbines belonging to the same wind field sector as the wind turbine, to obtain the target unit identifier with the strongest spatial correlation in the corresponding wind field sector, the target early warning time length of the target unit identifier at the wind turbine, and the actual spatial correlation coefficient.

[0113] In the present embodiment, because the wind directions of different wind field sectors are obviously different, the wind turbines in different wind field sectors will not be affected by the same wind speed mutation weather condition, and only the wind turbines belonging to the same wind field sector can be affected by the same wind speed mutation weather condition. On this basis, the computer device 10 does not need to perform cross-wind field sector sliding space correlation analysis on all wind turbines in the target wind farm, and only needs to perform sliding space correlation analysis on the wind turbines in each wind field sector, thereby effectively improving the model training efficiency of the target wind speed mutation early warning model.

[0114] Therefore, after obtaining the wind speed mutation sample set of each wind turbine (for example, wind turbine a, wind turbine b, …, wind turbine f) in a certain wind farm sector A, the computer device 10 performs a sliding spatial correlation analysis on each wind turbine (for example, wind turbine a) and other wind turbines (for example, wind turbine b, wind turbine c, …, wind turbine f) belonging to the same wind farm sector A according to the wind speed mutation sample set of the wind turbine (for example, wind turbine a) to ensure the actual analysis efficiency of the sliding spatial correlation analysis operation and to ensure that the final target wind turbine identification (for example, the wind turbine identification of wind turbine b) in the wind farm sector A with the strongest spatial correlation with the wind turbine a and the target early warning time length and the actual spatial correlation coefficient of the target wind turbine identification at the wind turbine a have obvious strength accuracy.

[0115] The step of performing a sliding spatial correlation analysis on each wind turbine (for example, wind turbine a) and other wind turbines (for example, wind turbine b, wind turbine c, …, wind turbine f) belonging to the same wind farm sector A according to the wind speed mutation sample set of the wind turbine (for example, wind turbine a) to obtain the target wind turbine identification (for example, the wind turbine identification of wind turbine b) in the corresponding wind farm sector with the strongest spatial correlation with the wind turbine and the target early warning time length and the actual spatial correlation coefficient of the target wind turbine identification at the wind turbine can include:

[0116] performing a sliding spatial correlation analysis on each wind speed mutation sample in the wind speed mutation sample set of the wind turbine and historical unit operation data of other wind turbines belonging to the same wind farm sector in a preset historical time period to obtain a spatial correlation matrix of the wind turbine at the wind speed mutation sample, wherein the spatial correlation matrix includes spearman correlation coefficients between the wind turbine and each other wind turbine at different sliding window time steps corresponding to the wind speed mutation sample;

[0117] taking the largest spearman correlation coefficient in the spatial correlation matrix of the wind turbine at the wind speed mutation sample as the optimal spatial correlation coefficient of the wind turbine at the wind speed mutation sample, and taking the sliding window time step corresponding to the optimal spatial correlation coefficient and the unit identification of the other wind turbine as the optimal early warning time length and the optimal strong correlation unit identification of the wind turbine at the wind speed mutation sample, respectively;

[0118] The optimal strong correlation unit identifier, the optimal early warning time length and the optimal spatial correlation coefficient corresponding to each wind speed mutation sample of the wind turbine are processed based on the principle of minority obeying majority, to obtain the target unit identifier with the strongest spatial correlation with the wind turbine, and the target early warning time length and the actual spatial correlation coefficient of the target unit identifier at the wind turbine.

[0119] Taking the existence of wind turbines a, b, …, f in the wind farm sector A as an example: if there are S wind speed mutation samples in the wind speed mutation sample set of wind turbine a, for the jth (j = 1, …, S) wind speed mutation sample of wind turbine a, the historical unit operation data of wind turbine a and wind turbines b, …, f in the wind farm sector A within a preset time period are required to be analyzed for sliding spatial correlation, to obtain the spearman correlation coefficient between wind turbine a and wind turbines b, …, f under different sliding window time steps (for example, T = 0, …, m) for the jth wind speed mutation sample of wind turbine a, and then all the spearman correlation coefficients related to the jth wind speed mutation sample are aggregated into a spatial correlation matrix Then the maximum spearman correlation coefficient in the spatial correlation matrix corresponding to the jth wind speed mutation sample of wind turbine a is extracted as the optimal spatial correlation coefficient M j of wind turbine a at the jth wind speed mutation sample j The sliding window time step T corresponding to the optimal spatial correlation coefficient M j and the unit identifier tagID of other wind turbines are respectively taken as the optimal early warning time length T j and the optimal strong correlation unit identifier tagID j of wind turbine a at the jth wind speed mutation sample.

[0120] Then, the optimal spatial correlation coefficient M j (j = 1, …, S), the optimal early warning time length T j (j = 1, …, S) and the optimal strong correlation unit identifier tagID pre (j = 1, …, S) corresponding to each wind speed mutation sample of wind turbine a are subjected to data analysis and processing under the principle of minority obeying majority, to obtain the target unit identifier (for example, the unit identifier tagID pre of wind turbine b) with the strongest spatial correlation with wind turbine a in the wind farm sector A, and the target early warning time length T pre and the actual spatial correlation coefficient M pre of the target unit identifier at wind turbine a.

[0121] Therefore, by performing the specific steps of step S260 above, this application can directly utilize the wind speed change sample set involving wind speed change conditions based on the sliding spatial correlation analysis operation to quickly and accurately mine the spatial correlation between multiple wind turbines in different wind farm sectors of the target wind farm, as well as the temporal correlation between multiple wind turbines in the same wind farm sector being affected by the same wind speed change conditions.

[0122] Step S270: Based on the wind speed change characteristic calculation strategy and the historical unit operation data of the wind turbine in the preset historical time period, construct a wind speed change early warning model of the wind turbine that matches the target unit identifier, the target early warning duration and the actual spatial correlation coefficient.

[0123] In this embodiment, when the computer device 10 discovers the target turbine identifier (tagID) of a wind turbine a within the target wind farm and the target wind turbine b within its respective wind farm sector A, which has the strongest spatial correlation with that of the target wind turbine a. pre and target unit identifier tagID pre The target early warning time T at point a of the wind turbine is... pre Correlation coefficient M with actual space pre Subsequently, the target unit identifier tagID corresponding to wind turbine unit a can be used. pre Target early warning duration T pre and the actual spatial correlation coefficient M pre Using a wind speed change feature calculation strategy and historical unit operation data of wind turbine a within a preset historical time period, a unit wind speed change early warning model for the target wind turbine a is constructed, which is related to the actual wind conditions of wind turbine a. The unit wind speed change early warning model provides wind speed change early warning for the target wind turbine a indicated by the target unit identifier.

[0124] Wherein, after the computer device 10 constructs a wind speed change early warning model for a wind turbine (e.g., wind turbine a mentioned above), it will perform model performance verification on the wind speed change early warning model according to preset model performance verification requirements (e.g., preset model iteration generation number and / or preset model accuracy). If the model performance verification fails, it will jump to step S260 above for the wind turbine and continue to execute until the constructed wind speed change early warning model for the wind turbine meets the preset model performance verification requirements, thereby ensuring that the accuracy of the early warning result of the wind speed change early warning model of each wind turbine in the target wind farm reaches the current best state.

[0125] Step S280, the corresponding unit wind speed mutation early warning model of each wind turbine of the target wind farm is subjected to model integration processing, and a target wind speed mutation early warning model matched with the target wind farm is obtained.

[0126] Therefore, the application can dig the spatial correlation conditions between multiple wind turbines in different wind farm sectors of the target wind farm and the time sequence correlation conditions of multiple wind turbines in the same wind farm sector affected by the same wind speed mutation wind condition by executing the above steps S250-S280, and construct a target wind speed mutation early warning model for ensuring that the wind speed mutation early warning effect of the target wind farm reaches the best state based on the dug spatial correlation conditions and time sequence correlation conditions.

[0127] In the application, in order to ensure that the computer device 10 can execute the above wind farm wind speed mutation early warning method through the wind farm wind speed mutation early warning device 100, the application realizes the above functions by function module division of the wind farm wind speed mutation early warning device 100. The specific composition of the wind farm wind speed mutation early warning device 100 provided by the application is described below.

[0128] Please refer to Figure 4 , Figure 4 is one of the composition schematic diagrams of the wind farm wind speed mutation early warning device 100 provided by the embodiments of the application. In the embodiments of the application, the wind farm wind speed mutation early warning device 100 can include an operating parameter acquisition module 110, a wind speed mutation detection module 120, an early warning parameter extraction module 130, and a mutation early warning module 140.

[0129] The operating parameter acquisition module 110 is configured to acquire actual unit operating data of each wind turbine in the target wind farm at the current time.

[0130] The wind speed mutation detection module 120 is configured to, for each wind turbine in the target wind farm, based on the actual unit operating data of the wind turbine, call the target wind speed mutation early warning model matched with the target wind farm to detect whether a wind speed mutation wind condition occurs at the wind turbine at the current time.

[0131] The early warning parameter extraction module 130 is configured to, in the case that the target wind speed mutation early warning model is called to detect that a wind speed mutation wind condition occurs at the wind turbine at the current time, extract, from the target wind speed mutation early warning model, a target unit identifier with the strongest spatial correlation with the wind turbine and a target early warning time length of the target unit identifier at the wind turbine.

[0132] The mutation early warning module 140 is configured to send a wind speed mutation early warning message to a target wind turbine identified by the target group identifier, where the wind speed mutation early warning message is used to warn the target wind turbine of a wind speed mutation condition at a future time interval of the target early warning time length from a current time.

[0133] Optionally, referring to Figure 5 , Figure 5 is a second constituent diagram of the wind farm wind speed mutation early warning device 100 provided by the embodiment of the application. In the embodiment of the application, the wind farm wind speed mutation early warning device 100 can further include a mutation sample acquisition module 150, a spatial correlation analysis module 160, a warning model construction module 170, and a warning model integration module 180.

[0134] The mutation sample acquisition module 150 is configured to acquire a plurality of wind speed mutation sample sets of a plurality of wind turbines included in each wind field sector of the target wind farm in a preset historical time period, where each wind speed mutation sample set includes wind speed mutation samples corresponding to different historical wind speed mutation times of a corresponding wind turbine in the preset historical time period, each wind speed mutation sample includes historical turbine operation data of the corresponding wind turbine near the corresponding historical wind speed mutation time, and historical wind speed mutation features of the corresponding wind turbine calculated based on a wind speed mutation feature calculation strategy at the corresponding historical wind speed mutation time.

[0135] The spatial correlation analysis module 160 is configured to, for each wind turbine in the target wind farm, perform a sliding spatial correlation analysis on the wind turbine and other wind turbines belonging to the same wind field sector as the wind turbine according to the wind speed mutation sample set of the wind turbine, to obtain a target group identifier of a wind turbine with the strongest spatial correlation in a corresponding wind field sector, a target early warning time length of the target group identifier at the wind turbine, and an actual spatial correlation coefficient.

[0136] The warning model construction module 170 is configured to, based on the target group identifier, the target early warning time length, and the actual spatial correlation coefficient corresponding to the wind turbine, construct a turbine wind speed mutation early warning model of the wind turbine matching the target group identifier according to the wind speed mutation feature calculation strategy and historical turbine operation data of the wind turbine in the preset historical time period, where the turbine wind speed mutation early warning model is used to perform wind speed mutation early warning on a target wind turbine pointed to by the target group identifier.

[0137] The warning model integration module 180 is configured to perform model integration processing on the turbine wind speed mutation early warning models corresponding to all wind turbines of the target wind farm, to obtain a target wind speed mutation early warning model matching the target wind farm.

[0138] It should be noted that the wind farm wind speed mutation early warning device 100 provided by the embodiments of the present application has the same basic principle and technical effects as the wind farm wind speed mutation early warning method. For brevity, the description of the embodiments not mentioned in the part can refer to the description of the wind farm wind speed mutation early warning method.

[0139] In the embodiments provided by the present application, it should be understood that the disclosed device and method can also be implemented in other ways. The device embodiments described above are only schematic. For example, the flowcharts and block diagrams in the accompanying drawings show the possible implementation architectures, functions and operation of the device, method and computer program product according to the embodiments of the present application. In this regard, each block in the flowcharts or block diagrams can represent a module, a program segment or a part of code, which contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions noted in the blocks can occur in a different order than that noted in the accompanying drawings. For example, two consecutive blocks can actually be executed substantially in parallel, and sometimes they can be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and the combination of blocks in the block diagrams and / or flowcharts, can be implemented by a dedicated hardware-based system for executing the specified functions or actions, or can be implemented by a combination of dedicated hardware and computer instructions.

[0140] In addition, the functional modules in each of the embodiments of the present application can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part. If the functions are realized in the form of software function modules and sold or used as independent products, they can be stored in a readable storage medium. Based on this understanding, the technical solutions of the present application can be embodied in the form of a software product, which is stored in a storage medium and includes a number of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media that can store program codes.

[0141] In summary, in the wind farm wind speed mutation early warning method and device and the computer equipment provided in the application, after obtaining the actual unit operation data of all wind turbines in the target wind farm at the current time, for each wind turbine, based on the actual unit operation data of the wind turbine, the target wind speed mutation early warning model matched with the target wind farm is called to detect whether the wind speed mutation wind condition occurs at the wind turbine at the current time, and when it is detected that the wind speed mutation wind condition occurs at the wind turbine at the current time, the target unit identifier with the strongest spatial correlation of the wind turbine is extracted from the target wind speed mutation early warning model, and the target early warning time length of the target unit identifier at the wind turbine, then a wind speed mutation early warning message for warning the target wind turbine that the target wind turbine encounters the wind speed mutation wind condition at a future time interval of the target early warning time length from the current time is sent to the target wind turbine pointed by the target unit identifier, so that the spatial correlation relationship between different wind turbines in the target wind farm is utilized to early warn the wind speed mutation of each wind turbine, and the unit operation conditions of different wind turbines in the whole target wind farm are controlled to the state substantially matched with the actual wind condition of the target wind farm, so as to improve the operation stability and power generation efficiency of each wind turbine in the target wind farm.

[0142] The above is only various embodiments of the application, but the protection scope of the application is not limited thereto, any person skilled in the art can easily think of changes or replacements within the technical range disclosed in the application, which should be covered in the protection scope of the application. Therefore, the protection scope of the application should be subject to the protection scope of the claims.

Claims

1. A method for early warning of sudden changes in wind speed in a wind farm, characterized in that, The method includes: Obtain the actual unit operation data of all wind turbines in the target wind farm at the current moment; For each wind turbine in the target wind farm, based on the actual unit operation data of the wind turbine, the target wind speed change early warning model matched with the target wind farm is invoked to detect whether a sudden wind speed change has occurred at the wind turbine. When the target wind speed change early warning model is called and a wind speed change occurs at the wind turbine, the target turbine identifier with the strongest spatial correlation with the wind turbine and the target early warning duration of the target turbine identifier at the wind turbine are extracted from the target wind speed change early warning model. A wind speed change warning message is sent to the target wind turbine unit identified by the target unit within the target wind farm. The wind speed change warning message is used to warn the target wind turbine unit that it will encounter a wind speed change at a future time with an interval of the target advance warning time from the current time.

2. The method according to claim 1, characterized in that, The step of using the actual operating data of the wind turbine to call a target wind speed change early warning model matched with the target wind farm to detect whether a sudden wind speed change has occurred at the wind turbine location includes: Based on the wind speed change feature calculation strategy recorded by the target wind speed change early warning model, the current wind speed change feature of the wind turbine corresponding to the actual unit operation data is calculated. Calculate the difference in abrupt changes in wind speed between the current wind turbine and multiple historical wind speed abrupt changes in wind turbine within a historical time period close to the current moment. The system checks whether the calculated differences in multiple abrupt change features satisfy the pre-stored boundary conditions for various wind speed abrupt changes. If the detected difference between the multiple abrupt change features satisfies at least one boundary condition for a sudden change in wind speed, it is determined that a sudden change in wind speed has occurred at the wind turbine.

3. The method according to claim 1, characterized in that, The method further includes: Obtain wind speed change sample sets for each of the multiple wind turbine units in different wind farm sectors of the target wind farm within a preset historical time period. Each wind speed change sample set includes wind speed change samples corresponding to different historical wind speed change times of the corresponding wind turbine unit within the preset historical time period. Each wind speed change sample includes historical turbine operation data of the corresponding wind turbine unit near the corresponding historical wind speed change time, and historical wind speed change features of the corresponding wind turbine unit calculated based on the wind speed change feature calculation strategy at the corresponding historical wind speed change time. For each wind turbine in the target wind farm, based on the wind speed change sample set of the wind turbine, a sliding space correlation analysis is performed on the wind turbine and other wind turbines belonging to the same wind farm sector as the wind turbine to obtain the target turbine identifier with the strongest spatial correlation with the wind turbine in the corresponding wind farm sector, as well as the target early warning time and actual spatial correlation coefficient of the target turbine identifier at the wind turbine. Based on the wind speed change feature calculation strategy and the historical unit operation data of the wind turbine in the preset historical time period, a wind speed change early warning model matching the target unit identifier corresponding to the wind turbine is constructed based on the target unit identifier, the target early warning duration and the actual spatial correlation coefficient. The wind speed change early warning model is used to provide wind speed change early warning for the target wind turbine pointed to by the target unit identifier. The wind speed change early warning models corresponding to each wind turbine in the target wind farm are integrated to obtain a target wind speed change early warning model that matches the target wind farm.

4. The method according to claim 3, characterized in that, The step of obtaining a sample set of wind speed change data for each wind turbine unit in different wind farm sectors of the target wind farm within a preset historical time period includes: For each wind turbine in the target wind farm, acquire the historical turbine operation data of each wind turbine at all historical moments within the preset historical time period; Based on the wind speed mutation feature calculation strategy, the historical wind speed mutation features of the wind turbine at different historical times are calculated, corresponding to the historical unit operation data. For each historical moment within the preset historical time period, detect whether the wind turbine exhibits a sudden change in wind speed at the corresponding historical moment. If a wind speed change is detected in the historical wind speed change characteristics of the wind turbine at a corresponding historical moment, the historical moment is taken as a historical wind speed change moment of the wind turbine, and the historical turbine operation data near the historical wind speed change moment and the historical wind speed change characteristics of the wind turbine at the historical wind speed change moment are integrated into a wind speed change sample of the wind turbine.

5. The method according to claim 4, characterized in that, The step of detecting whether the wind turbine exhibits a sudden wind speed change in its historical wind speed characteristics at a corresponding historical moment includes: Calculate the difference in abrupt changes in historical wind speed at the corresponding historical moment between the abrupt changes in historical wind speed at the corresponding historical moment and between the abrupt changes in historical wind speed at multiple historical time periods close to the corresponding historical moment. The system checks whether the calculated differences in multiple abrupt change features satisfy the pre-stored boundary conditions for various wind speed abrupt changes. If the difference between the multiple abrupt change features is found to satisfy at least one wind speed change boundary condition, it is determined that the wind turbine has experienced a wind speed change in the historical wind speed change feature at the corresponding historical time.

6. The method according to claim 3, characterized in that, For each wind turbine in the target wind farm, the step of performing a sliding space correlation analysis on the wind turbine and other wind turbines belonging to the same wind farm sector based on the wind speed change sample set of the wind turbine, to obtain the target turbine identifier with the strongest spatial correlation with the wind turbine in the corresponding wind farm sector, and the target early warning duration and actual spatial correlation coefficient of the target turbine identifier at the wind turbine location, includes: For each wind speed mutation sample in the wind speed mutation sample set of the wind turbine, a sliding spatial correlation analysis is performed on the historical unit operation data of the wind speed mutation sample and other wind turbines belonging to the same wind farm sector as the wind turbine within the preset historical time period to obtain the spatial correlation matrix of the wind turbine at the wind speed mutation sample. The spatial correlation matrix includes the Spearman correlation coefficient between the corresponding wind speed mutation sample of the wind turbine and each other wind turbine under different sliding window time steps. The Spearman correlation coefficient with the largest value in the spatial correlation matrix of the wind turbine at the wind speed change sample is taken as the optimal spatial correlation coefficient of the wind turbine at the wind speed change sample. The sliding window time step corresponding to the optimal spatial correlation coefficient and the unit identifier of other wind turbines are respectively taken as the optimal early warning duration and the optimal strongly correlated unit identifier of the wind turbine at the wind speed change sample. Based on the principle of majority rule, data processing is performed on the optimal strongly correlated turbine identifier, optimal early warning duration, and optimal spatial correlation coefficient corresponding to each wind speed change sample of the wind turbine. This yields the target turbine identifier with the strongest spatial correlation to the wind turbine, as well as the target early warning duration and actual spatial correlation coefficient of the target turbine identifier at the wind turbine location.

7. The method according to any one of claims 2-6, characterized in that, The strategy for calculating the abrupt change in wind speed is represented by a function expression that satisfies the continuity of the function, as shown below: f(Ω i ,V t ,b t ,W t )=ω1*V t +ω2*β t +ω3*W t ; Where, f(Ω) i V t ,β t W t ) is used to represent the wind speed change characteristics of the i-th wind turbine at time t, Ω i V is used to represent the nonlinear gain vector of the i-th wind turbine that adapts to the wind power curve. t β is used to represent the wind speed magnitude included in the unit operation data of the i-th wind turbine at time t. t W is used to represent the blade angle size included in the unit operation data of the i-th wind turbine at time t. t Ω is used to represent the power generation of the i-th wind turbine at time t, where Ω i = [ω1, ω2, ω3], where ω1, ω2 and ω3 are used to represent the actual nonlinear gain of the i-th wind turbine unit that is adapted to the wind power curve.

8. A wind speed sudden change early warning device for wind farms, characterized in that, The device includes: The operating parameter acquisition module is used to acquire the actual unit operating data of all wind turbines in the target wind farm at the current moment; The wind speed sudden change detection module is used to detect whether a sudden wind speed condition has occurred at each wind turbine in the target wind farm, based on the actual unit operation data of the wind turbine. The early warning parameter extraction module is used to extract the target turbine identifier with the strongest spatial correlation with the wind turbine and the target early warning duration of the target turbine identifier at the wind turbine when the target wind speed change early warning model is called and the wind speed change early warning model is detected. The sudden change early warning module is used to send a wind speed sudden change warning message to the target wind turbine unit identified by the target unit within the target wind farm. The wind speed sudden change warning message is used to warn the target wind turbine unit that it will encounter a sudden wind speed situation at a future time with an interval of the target early warning time from the current time.

9. The apparatus according to claim 8, characterized in that, The device further includes: The mutation sample acquisition module is used to acquire wind speed mutation sample sets of multiple wind turbines in different wind farm sectors of the target wind farm within a preset historical time period. Each wind speed mutation sample set includes wind speed mutation samples corresponding to different historical wind speed mutation times of the corresponding wind turbine within the preset historical time period. Each wind speed mutation sample includes historical turbine operation data of the corresponding wind turbine near the corresponding historical wind speed mutation time, and historical wind speed mutation features of the corresponding wind turbine calculated based on the wind speed mutation feature calculation strategy at the corresponding historical wind speed mutation time. The spatial correlation analysis module is used to perform sliding spatial correlation analysis on each wind turbine in the target wind farm, based on the wind speed change sample set of the wind turbine, and on other wind turbines belonging to the same wind farm sector as the wind turbine, to obtain the target turbine identifier with the strongest spatial correlation with the wind turbine in the corresponding wind farm sector, as well as the target early warning time and actual spatial correlation coefficient of the target turbine identifier at the wind turbine location. The early warning model construction module is used to construct a wind speed change early warning model for the wind turbine that matches the target turbine identifier based on the wind speed change feature calculation strategy and the historical turbine operation data of the wind turbine within the preset historical time period, and based on the target turbine identifier, the target early warning duration and the actual spatial correlation coefficient. The wind speed change early warning model is used to provide wind speed change early warning for the target wind turbine pointed to by the target turbine identifier. The early warning model integration module is used to perform model integration processing on the early warning models of wind speed change of each wind turbine in the target wind farm, so as to obtain a target wind speed change early warning model that matches the target wind farm.

10. A computer device, characterized in that, The method includes a processor and a memory, the memory storing a computer program that can be executed by the processor, the processor executing the computer program to implement the wind speed change early warning method for wind farms as described in any one of claims 1-7.

Images