Intelligent early warning method and system for wind farm based on running state monitoring

Abstract

The application relates to a wind farm intelligent early warning method and system based on operation state monitoring, which comprises the following steps: acquiring current external wind data of a wind farm and current operation state data of wind turbine generators in the wind farm, generating a current cabin estimated disturbance range; acquiring operation vibration detection data and operation noise detection data in the cabin of the wind turbine generator, generating a current operation estimated disturbance value, judging whether the current operation estimated disturbance value belongs to the current cabin estimated disturbance range; if the judgment is no, generating a current operation early warning indication, and guiding wind turbine generator maintenance personnel to maintain the cabin of the wind turbine generator according to the current operation early warning indication. The application realizes the monitoring and early warning of the cabins of wind turbine generators in the wind farm based on external wind data, operation state data and actual detection data in the cabin, realizes the early warning of possible faults, and reduces the failure rate.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of data acquisition, in particular to a wind farm intelligent early warning method and system based on operation state monitoring. BACKGROUND

[0002] The wind farm, i.e. the wind power generation field, is a field in which multiple large-scale grid-connected wind power generators are installed in a site with good wind energy resources, arranged in an array according to the terrain and the main wind direction, and combined to form a group to supply power to the power grid. According to different sites, the wind farm can be divided into an offshore wind farm and a land-based wind farm.

[0003] For the land-based wind farm, it is crucial to monitor its operation process. For example, an application for a kind of system and method for monitoring the operation state of a wind farm central monitoring system is disclosed in an invention patent application with the publication date of September 2, 2022 and the publication number of CN115002003A. The system includes a wind farm internal network state monitoring system and a wind farm external network state monitoring system for monitoring the wind farm central monitoring system. The wind farm internal network state monitoring system and the wind farm external network state monitoring system are respectively in communication connection with the centralized control Ethernet network.

[0004] Although the technical solution in the above patent document can monitor the operation state of the wind farm central monitoring system, judge possible problems of the system in advance, and avoid large-scale failure conditions such as system collapse, the monitoring of the wind farm is not comprehensive enough. Similar to other monitoring methods in the prior art, they all judge whether a fault is likely to occur by only focusing on the operation parameters themselves. However, there are often parts damaged due to vibration disturbance and other factors, but the damage does not affect the operation. At this time, if the fault warning is only based on the operation parameters, the warning will be delayed. Often, when the warning is given, the component has already been completely damaged, which may cause a major failure. SUMMARY

[0005] Therefore, it is necessary to provide a wind farm intelligent early warning method and system based on operation state monitoring, which can comprehensively consider the disturbance influence of external wind interference and vibration interference caused by normal operation on the cabin, and realize the monitoring and early warning of the cabin of each wind turbine generator in the wind farm based on external wind data, operation state data and actual detection data in the cabin.

[0006] The technical solution of the application is as follows:

[0007] A wind farm intelligent early warning method based on operation state monitoring, the method comprising:

[0008] obtaining current external wind data of the wind farm and current operation state data of the wind turbine generator in the wind farm, and generating a current cabin estimated disturbance range according to the current external wind data and the current operation state data;

[0009] obtaining operation vibration detection data and operation noise detection data in a nacelle of the wind turbine generator, and generating a current operation estimated disturbance value according to the operation vibration detection data and the operation noise detection data, wherein the current operation estimated disturbance value is a value for indicating that the nacelle is affected by operation of the wind turbine generator and external wind force;

[0010] judging whether the current operation estimated disturbance value belongs to the current nacelle estimated disturbance range;

[0011] if the judgment is no, generating a current operation early warning indication, and guiding a wind turbine generator maintenance personnel to maintain the nacelle of the wind turbine generator according to the current operation early warning indication.

[0012] Optionally, the nacelle comprises a plurality of vibration detection regions, and each vibration detection region is provided with a vibration detection point, and each vibration detection point is provided with a vibration detection sensor, and the operation vibration detection data comprises a sum of data detected by the activated vibration detection sensors, wherein one detection point operation vibration data is obtained after vibration detection by one activated vibration detection sensor;

[0013] The nacelle further comprises a noise detection region, and the noise detection region is provided with a plurality of noise collection sensors, and the noise collection sensors are used for noise collection;

[0014] obtaining operation vibration detection data and operation noise detection data in a nacelle of the wind turbine generator, and generating a current operation estimated disturbance value according to the operation vibration detection data and the operation noise detection data, comprising:

[0015] obtaining a current wind force level corresponding to the current external wind force data, and generating a detection point activation number according to the current wind force level;

[0016] selecting a to-be-activated detection point from each vibration detection point according to the detection point activation number;

[0017] controlling a vibration detection sensor corresponding to each to-be-activated detection point to start and obtaining detection point operation vibration data detected by each activated vibration detection sensor;

[0018] generating an operation vibration interference value according to each detection point operation vibration data;

[0019] calculating a distance between a noise collection sensor and each to-be-activated detection point according to each to-be-activated detection point, and setting a to-be-activated noise collection sensor, wherein the selected noise collection sensor is a to-be-started noise sensor;

[0020] activating the noise collection sensor to be activated and acquiring running noise detection data;

[0021] generating a running noise interference value according to the running noise detection data;

[0022] adding the running vibration interference value and the running noise interference value and generating a current running estimated disturbance value.

[0023] Optionally, the to-be-activated detection points are selected from the vibration detection points according to the detection point activation number, including:

[0024] acquiring historical maintenance data of the vibration detection areas in the engine room according to the detection point activation number;

[0025] extracting historical maintenance quantities of the vibration detection areas from the historical maintenance data, arranging the historical maintenance quantities corresponding to the vibration detection areas in descending order, and generating a maintenance frequency set, wherein one vibration detection area corresponds to one historical maintenance quantity;

[0026] selecting vibration detection areas with the same detection point activation number from the maintenance frequency set and setting the vibration detection areas as to-be-activated detection areas;

[0027] selecting corresponding to-be-activated detection points according to the to-be-activated detection areas.

[0028] Optionally, the running vibration interference value is generated according to the detection point running vibration data, including:

[0029] calculating a detection point fluctuation value according to the detection point running vibration data, wherein one detection point fluctuation value corresponds to one detection point running vibration data;

[0030] adding the detection point fluctuation values and generating the running vibration interference value.

[0031] Optionally, one detection point fluctuation value is calculated according to one detection point running vibration data, including:

[0032] calculating a maximum vibration frequency, a minimum vibration frequency and an average vibration frequency according to the detection point running vibration data;

[0033] generating a vibration frequency distribution curve according to the detection point running vibration data, and selecting an ascending frequency segment and a descending frequency segment from the vibration frequency distribution curve;

[0034] calculating slopes of the ascending frequency segment and the descending frequency segment, and taking absolute values of the slopes as fluctuation slopes;

[0035] generating a detection point fluctuation value according to the maximum vibration frequency, the minimum vibration frequency, the average vibration frequency and the fluctuation slopes.

[0036] Optionally, the distance between each of the noise collection sensors and the to-be-activated detection point is calculated according to each of the to-be-activated detection points, and the to-be-activated noise collection sensor is set, including:

[0037] The distance between each of the noise collection sensors and the to-be-activated detection point is calculated, and a sensor distance data set is generated, wherein one of the noise collection sensors corresponds to one of the sensor distance data sets, and each of the sensor distance data sets includes a plurality of actual detection point distances, and the distance between one of the noise collection sensors and one of the to-be-activated detection points is one of the actual detection point distances.

[0038] The sum of the distances between each of the noise collection sensors and each of the to-be-activated detection points is calculated according to each of the sensor distance data sets, and a total detection point distance is generated, wherein one of the noise collection sensors corresponds to one of the total detection point distances.

[0039] The noise collection sensor corresponding to the total detection point distance with the smallest value is set as the to-be-activated noise collection sensor.

[0040] Optionally, the running noise interference value is generated according to the running noise detection data, including:

[0041] The noise data frequency curve is generated according to the running noise detection data.

[0042] The total length of the noise data frequency curve is obtained, and an abnormal frequency curve is screened out from the noise data frequency curve, wherein the abnormal frequency curve is a curve outside the normal frequency range in the noise data frequency curve.

[0043] The abnormal curve length of the abnormal frequency curve is calculated.

[0044] The ratio of the abnormal curve length to the total length of the curve is calculated, and a running noise interference value is generated.

[0045] Optionally, the current external wind data of the wind farm and the current running state data of the wind turbine generator in the wind farm are obtained, and a current cabin-estimated disturbance range is generated according to the current external wind data and the current running state data, including:

[0046] The current external wind data of the wind farm is obtained, and the current wind force level is extracted from the current external wind data, and the current wind force disturbance coefficient is obtained according to the current wind force level;

[0047] The used time of the cabin is obtained, and a disturbance increment coefficient is generated according to the used time;

[0048] acquire current operation state data of wind turbines in the wind farm, compare the current operation state data with preset normal operation state data, and screen normal operation state data matching the current operation state data;

[0049] acquire an initial disturbance range according to the normal operation state data;

[0050] generate a current cabin estimated disturbance range according to the current wind disturbance coefficient, the disturbance amplitude coefficient, and the initial disturbance range.

[0051] Optionally, a wind farm intelligent early warning system based on operation state monitoring is also provided, and the system comprises:

[0052] an operation state data analysis module, configured to acquire current external wind data of the wind farm and current operation state data of wind turbines in the wind farm, and generate a current cabin estimated disturbance range according to the current external wind data and the current operation state data;

[0053] an operation disturbance generation module, configured to acquire operation vibration detection data and operation noise detection data in a cabin of the wind turbine, and generate a current operation estimated disturbance value according to the operation vibration detection data and the operation noise detection data, wherein the current operation estimated disturbance value is a value for indicating that the cabin is affected by operation of the wind turbine and external wind;

[0054] an operation disturbance judgment module, configured to judge whether the current operation estimated disturbance value belongs to the current cabin estimated disturbance range;

[0055] an operation early warning generation module, configured to generate a current operation early warning indication if the judgment is no, and guide wind turbine maintenance personnel to maintain the cabin of the wind turbine according to the current operation early warning indication.

[0056] Optionally, the cabin comprises a plurality of vibration detection regions, and each vibration detection region is provided with a vibration detection point, and each vibration detection point is provided with a vibration detection sensor, and the operation vibration detection data comprises a sum of data detected by activated vibration detection sensors, wherein one detection point operation vibration data is obtained after vibration detection by one activated vibration detection sensor;

[0057] The cabin further comprises a noise detection region, and the noise detection region is provided with a plurality of noise collection sensors, and the noise collection sensors are used for noise collection.

[0058] The operation disturbance generation module is further configured to:

[0059] obtaining a current wind level corresponding to the current external wind data, and generating a detection point activation number according to the current wind level; selecting to-be-activated detection points from the vibration detection points according to the detection point activation number; starting vibration detection sensors corresponding to the to-be-activated detection points and obtaining detection point operation vibration data detected by the activated vibration detection sensors; generating an operation vibration interference value according to the detection point operation vibration data; calculating distances between noise collection sensors and the to-be-activated detection points according to the to-be-activated detection points, and setting to-be-activated noise collection sensors, wherein the selected noise collection sensors are to-be-started noise sensors; starting the to-be-activated noise collection sensors and obtaining operation noise detection data; generating an operation noise interference value according to the operation noise detection data; and adding the operation vibration interference value and the operation noise interference value to generate a current operation estimated disturbance value.

[0060] Optionally, the operation disturbance generation module is further configured to:

[0061] obtaining historical maintenance data of the vibration detection regions in the cabin according to the detection point activation number; extracting historical maintenance numbers of the vibration detection regions from the historical maintenance data, arranging the historical maintenance numbers of the vibration detection regions in descending order, and generating a maintenance frequency set, wherein one vibration detection region corresponds to one historical maintenance number; selecting vibration detection regions with the same number of detection points as the detection point activation number from the maintenance frequency set, and setting the selected vibration detection regions as to-be-activated detection regions; and selecting to-be-activated detection points corresponding to the to-be-activated detection regions.

[0062] Optionally, the operation disturbance generation module is further configured to: calculate detection point fluctuation values according to the detection point operation vibration data, wherein one detection point fluctuation value corresponds to one detection point operation vibration data; add the detection point fluctuation values to generate the operation vibration interference value.

[0063] The operation disturbance generation module is further configured to: calculate the operation vibration interference value by the following formula:

[0064] ;

[0065] wherein Rd is the operation vibration interference value, is a vibration influence coefficient of an i-th vibration detection region, N is a number of detection point fluctuation values, and Fdi is an i-th detection point fluctuation value.

[0066] Optionally, the operation disturbance generation module is further configured to: calculate a maximum vibration frequency, a minimum vibration frequency and an average vibration frequency according to the vibration data of the detection point; generate a vibration frequency distribution curve according to the vibration data of the detection point, and screen an ascending frequency segment and a descending frequency segment from the vibration frequency distribution curve; calculate the slopes of the ascending frequency segment and the descending frequency segment, and take absolute values of the slopes as fluctuation slopes; and generate a detection point fluctuation value according to the maximum vibration frequency, the minimum vibration frequency, the average vibration frequency and the fluctuation slopes based on the following formula:

[0067] ;

[0068] wherein Fd is the detection point fluctuation value, is a first vibration coefficient, far is the average vibration frequency, and fs is a pre-stored standard vibration frequency, is a second vibration coefficient, fmax is the maximum vibration frequency, and fmin is the minimum vibration frequency, is a third vibration coefficient, n is the number of the fluctuation slopes, Kj is the jth fluctuation slope, and Ksj is a pre-stored standard slope.

[0069] The operation disturbance generation module is further configured to: calculate the distances between each of the noise collection sensors and the detection points to be activated, and generate sensor distance data sets, wherein one of the noise collection sensors corresponds to one of the sensor distance data sets, and each of the sensor distance data sets includes a plurality of actual detection point distances, and the distance between one of the noise collection sensors and one of the detection points to be activated is one of the actual detection point distances; calculate the total distances between each of the noise collection sensors and each of the detection points to be activated according to each of the sensor distance data sets, and generate detection point total distance sets, wherein one of the noise collection sensors corresponds to one of the detection point total distance sets; and set the noise collection sensor corresponding to the detection point total distance set with the smallest value as a noise collection sensor to be activated.

[0070] The operation disturbance generation module is further configured to: generate a noise data frequency curve according to the operation noise detection data; obtain a total length of the noise data frequency curve, and screen an abnormal frequency curve from the noise data frequency curve, wherein the abnormal frequency curve is a curve outside a normal frequency range in the noise data frequency curve; calculate an abnormal curve length of the abnormal frequency curve; calculate a ratio of the abnormal curve length to the total length of the noise data frequency curve, and generate an operation noise disturbance value.

[0071] The state data analysis module is further configured to: acquire current external wind data of the wind farm, extract a current wind level from the current external wind data, and acquire a current wind disturbance coefficient according to the current wind level; acquire a used time of the cabin, and generate a disturbance increase coefficient according to the used time; acquire current running state data of the wind turbine generator in the wind farm, compare the current running state data with preset conventional running state data, and screen out conventional running state data matched with the current running state data; acquire an initial disturbance range according to the conventional running state data; and generate a current cabin-estimated disturbance range according to the current wind disturbance coefficient, the disturbance increase coefficient, and the initial disturbance range.

[0072] Optionally, a computer device is also provided, which comprises a memory and a processor, the memory stores a computer program, and the processor implements the steps of the wind farm intelligent early warning method based on running state monitoring when executing the computer program.

[0073] Optionally, a computer readable storage medium is also provided, which stores a computer program, and the computer program implements the steps of the wind farm intelligent early warning method based on running state monitoring when executed by a processor.

[0074] The present application achieves the following technical effects:

[0075] The wind farm intelligent early warning method and system based on the operation state monitoring, in turn, obtains current external wind data of the wind farm and current operation state data of the wind turbine generator unit in the wind farm, and generates a current cabin estimated disturbance range according to the current external wind data and the current operation state data; obtains operation vibration detection data and operation noise detection data in the nacelle of the wind turbine generator unit, and generates a current operation estimated disturbance value according to the operation vibration detection data and the operation noise detection data, wherein the current operation estimated disturbance value is a value for indicating that the nacelle is affected by the operation of the wind turbine generator unit and external wind; it is judged whether the current operation estimated disturbance value belongs to the current cabin estimated disturbance range; if not, a current operation early warning indication is generated, and the wind turbine generator unit maintenance personnel is guided to maintain the nacelle of the wind turbine generator unit according to the current operation early warning indication. In the application, in order to judge whether the external wind of the wind farm and the operation data of the wind turbine generator unit will directly or indirectly affect the components in the nacelle of the wind turbine generator unit, the current external wind data of the wind farm and the current operation state data of the wind turbine generator unit in the wind farm are first obtained, and in the case that the operation of the wind turbine generator unit is normal, the disturbance in the nacelle caused by certain external wind and corresponding operation state will be within a normal range, which can be used as a reference for judging whether the operation of each component in the nacelle of the wind turbine generator unit will be affected, that is, a current cabin estimated disturbance range is generated according to the current external wind data and the current operation state data. That is, a current cabin estimated disturbance range is generated according to the current external wind data and the current operation state data, then operation vibration detection data and operation noise detection data in the nacelle of the wind turbine generator unit are obtained, and a current operation estimated disturbance value is generated according to the operation vibration detection data and the operation noise detection data, the current operation estimated disturbance value is set to represent the influence of the operation of the wind turbine generator unit and external wind on the nacelle, and the influence is represented by a specific numerical value, realizing data-based wind farm management, then it is judged whether the current operation estimated disturbance value belongs to the current cabin estimated disturbance range; if yes, that is, it is judged that the current operation estimated disturbance value belongs to the current cabin estimated disturbance range, at this time, it is indicated that no early warning is needed, therefore, a current operation safety indication is generated, and the wind turbine generator unit is controlled to continue operation according to the operation safety indication. If not, that is, it is judged that the current operation estimated disturbance value does not belong to the current cabin estimated disturbance range, a current operation early warning indication is generated, and the wind turbine generator unit maintenance personnel is guided to maintain the nacelle of the wind turbine generator unit according to the current operation early warning indication.Compared with the prior art, the present application considers the potential risks caused by the interference of the running and external wind on the cabin, and in the case of normal running parameters, some parts in the cabin may have been displaced or partially damaged. Such incomplete damage of parts does not affect normal operation, but if the time is too long, it will cause complete damage of parts, and further cause unit failure or other large loss. Therefore, the present application considers the disturbance influence of external wind interference and vibration interference caused by normal operation on the cabin, and realizes the monitoring and early warning of the cabin of each wind turbine in the wind farm based on external wind data, running state data and actual detection data in the cabin, realizes early warning of possible faults, and further reduces the failure rate. BRIEF DESCRIPTION OF DRAWINGS

[0076] Figure 1 A flowchart of a wind farm intelligent early warning method based on running state monitoring in an embodiment;

[0077] Figure 2 A structure block diagram of a wind farm intelligent early warning system based on running state monitoring in an embodiment. DETAILED DESCRIPTION

[0078] In order to make the purpose, technical scheme and advantages of the present application clearer, the present application will be further described in detail below with reference to the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application, and are not used to limit the present application.

[0079] In one embodiment, a terminal is provided, which is used to: acquire current external wind data of a wind farm and current running state data of a wind turbine in the wind farm, and generate a current cabin estimated disturbance range according to the current external wind data and the current running state data; acquire running vibration detection data and running noise detection data in a cabin of the wind turbine, and generate a current running estimated disturbance value according to the running vibration detection data and the running noise detection data, wherein the current running estimated disturbance value is a value for indicating that the cabin is affected by the running of the wind turbine and external wind; judge whether the current running estimated disturbance value belongs to the current cabin estimated disturbance range; if not, generate a current running early warning indication, and guide the wind turbine maintenance personnel to maintain the cabin of the wind turbine according to the current running early warning indication.

[0080] The terminal can be, but is not limited to, various personal computers, notebook computers, smart phones, tablet computers and portable wearable devices.

[0081] In one embodiment, asFigure 1 As shown, a wind farm intelligent early warning method based on running state monitoring is provided, and the method comprises:

[0082] Step S100: Obtain current external wind data of a wind farm and current running state data of wind turbines in the wind farm, and generate a current cabin estimated disturbance range according to the current external wind data and the current running state data;

[0083] Step S200: Obtain running vibration detection data and running noise detection data in a cabin of the wind turbine, and generate a current running estimated disturbance value according to the running vibration detection data and the running noise detection data, wherein the current running estimated disturbance value is a value for indicating that the cabin is affected by the running of the wind turbine and external wind;

[0084] Step S300: Determine whether the current running estimated disturbance value belongs to the current cabin estimated disturbance range;

[0085] Step S400: If the determination is no, generate a current running early warning indication, and guide a wind turbine maintenance personnel to maintain the cabin of the wind turbine according to the current running early warning indication.

[0086] In this embodiment, in order to determine whether the wind outside the wind farm and the operation data of the wind turbine generator will have a direct or indirect impact on the components in the nacelle of the wind turbine generator, the current external wind data of the wind farm and the current operation state data of the wind turbine generator in the wind farm are first obtained. When the wind turbine generator is operating normally, a certain external wind and the corresponding operation state will cause disturbance in the nacelle within a normal range. This range can be used as a reference for determining whether it will affect the operation of each component in the nacelle of the wind turbine generator, that is, a current in-cabin estimated disturbance range is generated according to the current external wind data and the current operation state data. That is, a current in-cabin estimated disturbance range is generated according to the current external wind data and the current operation state data. Then, the operation vibration detection data and the operation noise detection data in the nacelle of the wind turbine generator are obtained, and a current operation estimated disturbance value is generated according to the operation vibration detection data and the operation noise detection data. The current operation estimated disturbance value is used to represent the degree of influence of the nacelle caused by the operation of the wind turbine generator and the external wind, and is represented by a specific numerical value, realizing data-based wind farm management. Then, it is determined whether the current operation estimated disturbance value belongs to the current in-cabin estimated disturbance range. If the determination is yes, that is, it is determined that the current operation estimated disturbance value belongs to the current in-cabin estimated disturbance range, it means that no early warning is needed, therefore, a current operation safety indication is generated, and the wind turbine generator is controlled to continue operating according to the operation safety indication. If the determination is no, that is, it is determined that the current operation estimated disturbance value does not belong to the current in-cabin estimated disturbance range, a current operation early warning indication is generated, and the wind turbine generator maintenance personnel is guided to maintain the nacelle of the wind turbine generator according to the current operation early warning indication. Compared with the early warning method in the prior art which only analyzes whether a fault is likely to occur according to the operation parameters themselves, the present application considers the potential risks caused by the interference of the operation and the external wind on the nacelle. Usually, in the case of normal operation parameters, some components in the nacelle may have already been displaced or partially damaged. Such incomplete damage of components will not affect normal operation, but if the time is too long, it will cause complete damage of the components, and further cause unit failure or other large loss. Therefore, the present application considers the disturbance influence of the external wind interference and the vibration interference caused by normal operation on the nacelle, and realizes the monitoring and early warning of the nacelles of each wind turbine generator in the wind farm based on the external wind data, the operation state data and the actual detection data in the nacelle, realizes early warning of possible faults, and further reduces the failure rate.

[0087] In one embodiment, the cabin includes a plurality of vibration detection areas, each of which is provided with a vibration detection point, and one of the vibration detection points is provided with a vibration detection sensor, the operation vibration detection data includes the sum of the data detected by the activated vibration detection sensors, wherein one activated vibration detection sensor obtains one detection point operation vibration data after vibration detection;

[0088] The cabin also includes a noise detection area, and the noise detection area is provided with a plurality of noise collection sensors, and the noise collection sensors are used for noise collection;

[0089] Step S200: Obtain the operation vibration detection data and operation noise detection data in the cabin of the wind turbine generator, and generate a current operation estimated disturbance value according to the operation vibration detection data and the operation noise detection data, comprising:

[0090] Step S210: Obtain the current wind level corresponding to the current external wind data, and generate a detection point activation number according to the current wind level;

[0091] Step S220: Select an activated detection point from each of the vibration detection points according to the detection point activation number;

[0092] Step S230: Control the vibration detection sensor corresponding to each of the activated detection points to start and obtain the detection point operation vibration data detected by each of the activated vibration detection sensors;

[0093] Step S240: Generate an operation vibration interference value according to each of the detection point operation vibration data;

[0094] Step S250: Calculate the distance between the noise collection sensor and the activated detection point according to each of the activated detection points, and set an activated noise collection sensor, wherein the selected noise collection sensor is a noise sensor to be started;

[0095] Step S260: Start the activated noise collection sensor and obtain the operation noise detection data;

[0096] Step S270: Generate an operation noise interference value according to the operation noise detection data;

[0097] Step S280: Add the operation vibration interference value and the operation noise interference value to generate a current operation estimated disturbance value.

[0098] In this embodiment, in order to achieve all-around fine detection in the maximum interference condition, further by setting multiple vibration detection areas in the cabin, and making each vibration detection area be provided with a vibration detection point, one vibration detection point is provided with a vibration detection sensor, the running vibration detection data includes the data detected by the activated vibration detection sensor. Further, in order to prevent the redundancy of detection data and save monitoring resources, and considering that when the external wind force is small or the running power is low, the disturbance is small, so at this time, the full sensor starting control is not needed, that is, only a part of the vibration detection sensor needs to be activated. Therefore, one vibration detection sensor is set to perform vibration detection, and a detection point running vibration data is obtained. At the same time, by setting a noise detection area in the cabin, multiple noise collection sensors are arranged in the noise detection area, and the noise collection sensors are used for noise collection. Then, the current wind level corresponding to the current external wind data is obtained, and the detection point activation number is generated according to the current wind level. Specifically, after extracting the current wind level from the current external wind data, the current wind level is compared with the pre-stored standard wind level database, and the standard wind level database pre-stores multiple wind test groups, each of which includes a standard wind level and a standard detection point number. Therefore, the standard wind level same as the current wind level is screened out, and the corresponding standard detection point number is set as the detection point activation number. For example, if the detection point activation number is 3, then 3 vibration detection sensors need to be activated. Then, the to-be-activated detection points are selected from the vibration detection points according to the detection point activation number, and the vibration detection sensors corresponding to the to-be-activated detection points are started and the detection point running vibration data detected by the activated vibration detection sensors is obtained. Then, the running vibration interference value is generated according to the detection point running vibration data, which is used to represent the influence degree of the vibration generated by the wind turbine in the running process on the vibration detection area corresponding to each to-be-activated detection point in the cabin. Then, the noise monitoring points are set according to each to-be-activated detection point, which is to perform noise detection according to the vibration detection points, realize the linkage monitoring of the vibration and noise of the selected area, then the noise collection sensor is selected from the noise detection area according to the noise monitoring point, then the to-be-started noise sensor in the noise detection area is started and the running noise detection data is obtained; and the running noise interference value is generated according to the running noise detection data, which is used to represent the influence degree of the noise generated by the wind turbine in the running process on the noise detection area corresponding to each noise monitoring point in the cabin.Finally, the running vibration interference value and the running noise interference value are added and a current running estimated disturbance value is generated, so that noise and vibration in the cabin are considered for analysis to determine whether a fault problem is likely to occur, thereby achieving early warning.

[0099] Further, the vibration sensor can adopt a microphone, a piezoelectric ceramic sheet, an acceleration sensor or other sensors capable of acquiring vibration signals. The noise collection sensor can adopt a microphone.

[0100] In one embodiment, step S220: selecting to-be-activated detection points from the vibration detection points according to the detection point activation number, comprises:

[0101] Step S221: obtaining historical maintenance data of each vibration detection area in the cabin according to the detection point activation number;

[0102] Step S222: extracting historical maintenance numbers of each vibration detection area from the historical maintenance data, arranging the historical maintenance numbers corresponding to each vibration detection area in descending order, and generating a maintenance frequency set, wherein one vibration detection area corresponds to one historical maintenance number;

[0103] Step S223: selecting vibration detection areas with the same detection point activation number from the maintenance frequency set and setting them as to-be-activated detection areas;

[0104] Step S224: selecting corresponding to-be-activated detection points according to the to-be-activated detection areas.

[0105] In this embodiment, in order to prevent detection of more useless data, the selection of the to-be-activated detection point is needed. In the selection, the historical maintenance data of each vibration detection area in the cabin is obtained according to the number of activated detection points. The historical maintenance data is the maintenance data left by each vibration detection area in the past maintenance and early warning process. Then, the historical maintenance number of each vibration detection area is extracted from the historical maintenance data, the historical maintenance number corresponding to each vibration detection area is arranged in descending order, and a maintenance frequency set is generated. In the maintenance frequency set, the values of the first historical maintenance number to the last historical maintenance number decrease in turn, that is, the first historical maintenance number has the maximum value. Then, the vibration detection areas with the same number of activated detection points are selected from the maintenance frequency set, and are set as the to-be-activated detection area. For example, there are five historical maintenance numbers in the maintenance frequency set. When the number of activated detection points is three, three are selected from the maintenance frequency set. Then, the corresponding to-be-activated detection point is selected according to the to-be-activated detection area. In this way, when the wind force is maximum and the running interference is maximum, the vibration detection sensors in all vibration detection areas are controlled to detect. If it is not the maximum wind force and the maximum interference, all vibration detection sensors do not need to work at the same time. At this time, by steps S221 to S224, the detection area most prone to failure in the historical data is selected, so that only the area most prone to failure is detected when all vibration detection sensors are not turned on, which saves detection resources and maximizes the probability of finding fault points and trend prediction.

[0106] In one embodiment, step S240: generating a running vibration interference value according to each detection point running vibration data, comprising:

[0107] Step S241: calculating a detection point fluctuation value according to the detection point running vibration data, wherein one detection point fluctuation value corresponds to one detection point running vibration data;

[0108] Step S242: adding each detection point fluctuation value to generate the running vibration interference value.

[0109] In this embodiment, the running vibration interference value is calculated by the following formula:

[0110] ;

[0111] wherein Rd is the running vibration interference value, is the vibration influence coefficient of the i-th vibration detection area, N is the number of detection point fluctuation values, and Fdi is the i-th detection point fluctuation value.

[0112] In the embodiment, the vibration detection area includes a first wheel hub detection area, a second wheel hub detection area, a main shaft detection area, a gearbox detection area and a cabin wall detection area. The vibration detection sensor corresponding to the vibration detection point of the first wheel hub detection area is arranged on the outer side of the wheel hub; the vibration detection sensor corresponding to the vibration detection point of the second wheel hub detection area is arranged on the wheel hub and opposite to the vibration detection point of the first wheel hub detection area; the vibration detection sensor corresponding to the vibration detection point of the main shaft detection area is arranged on the outer side of the main shaft; the vibration detection sensor corresponding to the vibration detection point of the gearbox detection area is arranged on the gearbox; and the vibration detection sensor corresponding to the vibration detection point of the cabin wall detection area is arranged on the inner wall of the cabin. Obviously, different vibration detection areas are used to detect different components. Specifically, the vibration influence coefficients corresponding to the first wheel hub detection area, the second wheel hub detection area, the main shaft detection area, the gearbox detection area and the cabin wall detection area are respectively The vibration influence coefficients are pre-set according to the components in the vibration detection area.

[0113] In one embodiment, a detection point fluctuation value is calculated according to the vibration data of a detection point, including:

[0114] Step S2411: The maximum vibration frequency, the minimum vibration frequency and the average vibration frequency are calculated according to the vibration data of the detection point;

[0115] Step S2412: The vibration frequency distribution curve is generated according to the vibration data of the detection point, and the rising frequency segment and the falling frequency segment are selected from the vibration frequency distribution curve;

[0116] Step S2413: The slopes of the rising frequency segment and the falling frequency segment are calculated, and the absolute values of the slopes are taken as the fluctuation slopes;

[0117] Step S2414: The detection point fluctuation value is generated according to the maximum vibration frequency, the minimum vibration frequency, the average vibration frequency and the fluctuation slope based on the following formula:

[0118]

[0119] wherein Fd is the detection point fluctuation value, is the first vibration coefficient, far is the average vibration frequency, and fs is the pre-stored standard vibration frequency, is the second vibration coefficient, fmax is the maximum vibration frequency, and fmin is the minimum vibration frequency, is the third vibration coefficient, n is the number of fluctuation slopes, Kj is the jth fluctuation slope, and Ksj is the pre-stored standard slope.

[0120] Further, the calculation formula of each detection point fluctuation value is as follows:​

[0121] ;

[0122] Where Fdi is the fluctuation value at the i-th detection point. Let fari be the first vibration coefficient, fsi be the average vibration frequency corresponding to the vibration data of the i-th detection point, and fsi be the pre-stored standard vibration frequency corresponding to the vibration data of the i-th detection point. Let fmaxi be the second vibration coefficient, fmini be the maximum vibration frequency corresponding to the vibration data at the i-th detection point, and fmin be the minimum vibration frequency corresponding to the vibration data at the i-th detection point. is the third vibration coefficient, n is the number of fluctuation slopes, Kj is the j-th fluctuation slope, and Ksj is the standard slope corresponding to the vibration data of the ith detection point.

[0123] The first vibration coefficient, the second vibration coefficient, and the third vibration coefficient are all preset. The first vibration coefficient represents the proportion of average fluctuation, the second vibration coefficient represents the proportion of the difference between the maximum and minimum frequencies, and the third vibration coefficient represents the proportion of frequency variation.

[0124] In one embodiment, step S250: calculating the distance between the noise acquisition sensor and the detection point to be activated based on each of the detection points to be activated, and setting the noise acquisition sensor to be activated, includes:

[0125] Step S251: Calculate the distance between each noise acquisition sensor and the detection point to be activated, and generate a sensor spacing data group, wherein one noise acquisition sensor corresponds to one sensor spacing data group, each sensor spacing data group includes multiple actual detection point spacings, and the distance between one noise acquisition sensor and one detection point to be activated is one actual detection point spacing.

[0126] Step S252: Calculate the total distance between each noise acquisition sensor and each detection point to be activated based on each sensor spacing data group, and generate the total detection point spacing, wherein one noise acquisition sensor corresponds to one total detection point spacing;

[0127] Step S253: Set the noise acquisition sensor corresponding to the total spacing of the detection points with the smallest value as the noise acquisition sensor to be activated.

[0128] In this embodiment, the generation of the sensor spacing data set is realized by sequentially calculating the distance between each noise collection sensor and the to-be-activated detection point, so that one noise collection sensor corresponds to one sensor spacing data set, each sensor spacing data set includes a plurality of actual detection point spacings, the distance between one noise collection sensor and one to-be-activated detection point is an actual detection point spacing, then the sum of the distances between each noise collection sensor and each to-be-activated detection point is calculated according to each sensor spacing data set, and a total detection point spacing is generated, wherein one noise collection sensor corresponds to one total detection point spacing, and finally the noise collection sensor corresponding to the total detection point spacing with the smallest value is set as the to-be-activated noise collection sensor. This setting makes the to-be-activated noise collection sensor not fixed, but set according to the to-be-activated detection point, and the to-be-activated noise collection sensor is the one closest to each to-be-activated detection point, which can maximize the avoidance of data loss caused by long distance. Considering the cost, it is extremely difficult to set enough noise collection sensors, so steps S251 to S253 are used to set the noise collection sensor closest to each to-be-activated detection point among a limited number of noise collection sensors, thereby maximizing the high-precision noise data collection and improving the accuracy of data warning.

[0129] In one embodiment, step S270: generating a running noise interference value according to the running noise detection data, comprises:

[0130] Step S271: generating a noise data frequency curve according to the running noise detection data;

[0131] Step S272: obtaining the total length of the noise data frequency curve, and screening an abnormal frequency curve from the noise data frequency curve, wherein the abnormal frequency curve is a curve outside the normal frequency range in the noise data frequency curve;

[0132] Step S273: calculating the abnormal curve length of the abnormal frequency curve;

[0133] Step S274: calculating the ratio of the abnormal curve length to the total length of the curve, and generating a running noise interference value.

[0134] In this embodiment, the historical data collected by the to-be-activated noise collection sensor when each component corresponding to the to-be-activated detection point is working is extracted in advance, and normal frequency data of the wind turbine when the wind turbine is working normally is filtered according to the historical data, the normal frequency data including a normal frequency maximum value and a normal frequency minimum value, and the normal frequency maximum value and the normal frequency minimum value forming a normal frequency range. Therefore, a noise data frequency curve is generated according to the running noise detection data, a total length of the noise data frequency curve is obtained, and an abnormal frequency curve is filtered from the noise data frequency curve, wherein the abnormal frequency curve is a curve outside the normal frequency range in the noise data frequency curve. Then, an abnormal curve length of the abnormal frequency curve is calculated, and finally, a ratio of the abnormal curve length to the total length of the curve is calculated, and a running noise interference value is generated. It should be noted that,

[0135] On the other hand, when the running vibration interference value and the running noise interference value are added, different weight values can be set for the running vibration interference value and the running noise interference value respectively, so as to realize self-defined setting. For example, the running vibration interference value and the running noise interference value correspond to a first weight value and a second weight value respectively. When maintenance is needed, the vibration influence is less than the noise influence for maintenance workers, and at this time, the weight value corresponding to the running noise interference value is set to be greater than the weight value corresponding to the running vibration interference value.

[0136] In one embodiment, step S100: obtaining current external wind data of a wind farm and current running state data of a wind turbine in the wind farm, and generating a current cabin-estimated disturbance range according to the current external wind data and the current running state data, comprising:

[0137] Step S110: obtaining current external wind data of a wind farm, extracting a current wind level from the current external wind data, and obtaining a current wind disturbance coefficient according to the current wind level;

[0138] Step S120: obtaining the used time of the cabin, and generating a disturbance increment coefficient according to the used time;

[0139] Step S130: obtaining current running state data of a wind turbine in a wind farm, comparing the current running state data with preset conventional running state data, and filtering conventional running state data matched with the current running state data;

[0140] Step S140: obtaining an initial disturbance range according to the conventional running state data;

[0141] Step S150: generating a current cabin estimated disturbance range according to the current wind disturbance coefficient, the disturbance amplification coefficient and the initial disturbance range.

[0142] In this embodiment, current external wind data of the wind farm is acquired, and a current wind level is extracted from the current external wind data, the current wind level being a value representing the wind strength outside the cabin, the greater the current wind level, the greater the wind strength. The wind speed and wind direction are detected by a preset wind monitoring sensor, and the detected wind speed and wind direction are classified to generate the current wind level. The current wind level is compared with the standard wind level database, the standard wind level database pre-storing a plurality of wind test groups, each of the wind test groups including a standard wind level corresponding to different standard wind disturbance coefficients. The standard wind disturbance coefficient corresponding to the same standard wind level as the current wind level is set as the current wind disturbance coefficient. Then, the used time of the cabin is acquired, and a disturbance amplification coefficient is generated according to the used time, the used time being proportional to the disturbance amplification coefficient, the longer the used time, the greater the disturbance amplification coefficient. The disturbance amplification coefficient is obtained by dividing the used time by the rated total life of the cabin. Then, current operating state data of the wind turbine generator in the wind farm is acquired, and the current operating state data is compared with preset conventional operating state data, and the conventional operating state data matching the current operating state data is screened out. The conventional operating state data is pre-stored and has a plurality of quantities.

[0143] The working state of the wind turbine generator is pre-acquired, and the data of the corresponding cabin vibration detection area and noise detection area is acquired, and the acquired data is used to establish the corresponding relationship between the operating data and the cabin noise and vibration data.

[0144] Generally, for the same operating state, the data acquired in the cabin vibration detection area and the noise detection area is not exactly the same, therefore, after the data acquisition is completed, the data is summarized and analyzed, and after the analysis is completed, a range value is generated, for the same motion parameter, the vibration data and the noise data in the cabin are a range value, that is, a standard disturbance range, the standard disturbance range can be understood as the influence of the vibration generated by the wind turbine generator during operation on the cabin vibration.

[0145] Further, the standard disturbance range value can be generated in the same way as the current operating estimated disturbance value.

[0146] Therefore, the conventional operation state data matched with the current operation state data is screened out first, and then the standard disturbance range corresponding to the screened conventional operation state data is set as an initial disturbance range. Further, considering the current actual wind force and the reduction of anti-interference ability caused by the increase of use time, the initial disturbance range is multiplied by the sum of the current wind force disturbance coefficient and the disturbance amplitude coefficient to finally generate a current cabin estimated disturbance range.

[0147] Further, the current operation state data includes but is not limited to real-time power output, cumulative power generation, maximum power point tracking state, impeller speed, generator speed and generator temperature.

[0148] In one embodiment, as shown in Figure 2 The system further includes a state data analysis module configured to acquire current external wind force data of the wind farm and current operation state data of the wind turbine generator in the wind farm, and generate a current cabin estimated disturbance range according to the current external wind force data and the current operation state data.

[0149] The state data analysis module is further configured to acquire current external wind force data of the wind farm and current operation state data of the wind turbine generator in the wind farm, and generate a current cabin estimated disturbance range according to the current external wind force data and the current operation state data.

[0150] The operation disturbance generation module is further configured to acquire operation vibration detection data and operation noise detection data in the nacelle of the wind turbine generator, and generate a current operation estimated disturbance value according to the operation vibration detection data and the operation noise detection data, wherein the current operation estimated disturbance value is a value for indicating that the nacelle is affected by the operation of the wind turbine generator and the external wind force.

[0151] The operation disturbance judgment module is further configured to judge whether the current operation estimated disturbance value belongs to the current cabin estimated disturbance range.

[0152] The operation warning generation module is further configured to generate a current operation warning indication if the judgment is negative, and guide the wind turbine generator maintenance personnel to maintain the nacelle of the wind turbine generator according to the current operation warning indication.

[0153] In one embodiment, the nacelle includes a plurality of vibration detection regions, and each vibration detection region is provided with a vibration detection point. One vibration detection point is provided with a vibration detection sensor. The operation vibration detection data includes the sum of the data detected by the activated vibration detection sensors, wherein one vibration detection point operation vibration data is obtained after the activated vibration detection sensor performs vibration detection.

[0154] The nacelle further includes a noise detection region, and the noise detection region is provided with a plurality of noise collection sensors. The noise collection sensors are used for noise collection.

[0155] The operation disturbance generation module is further configured to:

[0156] obtain a current wind level corresponding to the current external wind data, and generate a detection point activation number according to the current wind level; select to-be-activated detection points from the vibration detection points according to the detection point activation number; control vibration detection sensors corresponding to the to-be-activated detection points to start and obtain detection point operation vibration data detected by the activated vibration detection sensors; generate an operation vibration interference value according to the detection point operation vibration data; calculate distances between noise collection sensors and the to-be-activated detection points according to the to-be-activated detection points, and set to-be-activated noise collection sensors, wherein the selected noise collection sensors are to-be-started noise sensors; start the to-be-activated noise collection sensors and obtain operation noise detection data; generate an operation noise interference value according to the operation noise detection data; and add the operation vibration interference value and the operation noise interference value to generate a current operation estimated disturbance value.

[0157] In one embodiment, the operation disturbance generation module is further configured to:

[0158] obtain historical maintenance data of the vibration detection regions in the cabin according to the detection point activation number; extract historical maintenance numbers of the vibration detection regions from the historical maintenance data, arrange the historical maintenance numbers of the vibration detection regions in descending order, and generate a maintenance frequency set, wherein one vibration detection region corresponds to one historical maintenance number; select vibration detection regions with the same number of activations as the detection point activation number from the maintenance frequency set, and set the selected vibration detection regions as to-be-activated detection regions; and select to-be-activated detection points corresponding to the to-be-activated detection regions.

[0159] The operation disturbance generation module is further configured to: calculate detection point fluctuation values according to the detection point operation vibration data, wherein one detection point fluctuation value corresponds to one detection point operation vibration data; add the detection point fluctuation values to generate the operation vibration interference value.

[0160] In one embodiment, the operation disturbance generation module is further configured to: calculate the operation vibration interference value by the following formula:

[0161] ;

[0162] wherein Rd is the operation vibration interference value, is a vibration influence coefficient of the ith vibration detection region, N is the number of detection point fluctuation values, and Fdi is the ith detection point fluctuation value.

[0163] In one embodiment, the running disturbance generation module is further configured to: calculate a maximum vibration frequency, a minimum vibration frequency and an average vibration frequency according to the running vibration data of the detection point; generate a vibration frequency distribution curve according to the running vibration data of the detection point, and screen out an ascending frequency segment and a descending frequency segment from the vibration frequency distribution curve; calculate the slopes of the ascending frequency segment and the descending frequency segment, and take the absolute values of the slopes as fluctuation slopes; and generate a detection point fluctuation value according to the maximum vibration frequency, the minimum vibration frequency, the average vibration frequency and the fluctuation slopes based on the following formula:

[0164] ;

[0165] wherein Fd is the detection point fluctuation value, is a first vibration coefficient, far is the average vibration frequency, and fs is a pre-stored standard vibration frequency, is a second vibration coefficient, fmax is the maximum vibration frequency, and fmin is the minimum vibration frequency, is a third vibration coefficient, n is the number of the fluctuation slopes, Kj is the jth fluctuation slope, and Ksj is a pre-stored standard slope.

[0166] In one embodiment, the running disturbance generation module is further configured to: calculate the distances between each of the noise collection sensors and the detection points to be activated, and generate sensor distance data sets, wherein one of the noise collection sensors corresponds to one of the sensor distance data sets, and each of the sensor distance data sets includes a plurality of actual detection point distances, and the distance between one of the noise collection sensors and one of the detection points to be activated is one of the actual detection point distances; calculate the total distances between each of the noise collection sensors and each of the detection points to be activated according to each of the sensor distance data sets, and generate detection point total distance sets, wherein one of the noise collection sensors corresponds to one of the detection point total distance sets; and set the noise collection sensor corresponding to the detection point total distance set with the smallest value as the noise collection sensor to be activated.

[0167] In one embodiment, the running disturbance generation module is further configured to: generate a noise data frequency curve according to the running noise detection data; obtain a total length of the noise data frequency curve, and screen out an abnormal frequency curve from the noise data frequency curve, wherein the abnormal frequency curve is a curve outside the normal frequency range in the noise data frequency curve; calculate an abnormal curve length of the abnormal frequency curve; calculate a ratio of the abnormal curve length to the total length of the noise data frequency curve, and generate a running noise disturbance value.

[0168] In one embodiment, the state data analysis module is further configured to: acquire current external wind data of the wind farm, extract a current wind level from the current external wind data, and acquire a current wind disturbance coefficient according to the current wind level; acquire a used time of the cabin, and generate a disturbance increment coefficient according to the used time; acquire current running state data of the wind turbine generator within the wind farm, compare the current running state data with preset regular running state data, and filter out regular running state data matching the current running state data; acquire an initial disturbance range according to the regular running state data; and generate a current cabin-estimated disturbance range according to the current wind disturbance coefficient, the disturbance increment coefficient, and the initial disturbance range.

[0169] In one embodiment, a computer device is provided, which includes a memory and a processor, the memory stores a computer program, and the processor implements the steps of the above-described wind farm intelligent early warning method based on running state monitoring when executing the computer program.

[0170] In one embodiment, a computer readable storage medium is also provided, which stores a computer program, and the computer program implements the steps of the above-described wind farm intelligent early warning method based on running state monitoring when executed by a processor.

[0171] Those skilled in the art can understand that all or part of the processes in the above-described embodiments can be completed by a computer program instructing related hardware, and the computer program can be stored in a non-volatile computer readable storage medium. When executed, the computer program can include the processes of the above-described embodiments. Any reference to memory, storage, database or other medium used in the embodiments provided by the present application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration but not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct RAM bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0172] Any combination of the technical features in the above embodiments can be made, and for the sake of brevity, not all possible combinations are described above, however, as long as the combination of the technical features does not exist in contradiction, it shall be considered within the scope of the present disclosure.

[0173] The above embodiments only express several implementation manners of the present application, and the description is relatively specific and detailed, but it shall not be understood as a limitation on the patent scope of the present application. It shall be pointed out that, for ordinary skilled persons in the art, several modifications and improvements can be made without departing from the concept of the present application, and these shall be within the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the appended claims.

Claims

1. A wind farm intelligent early warning method based on running state monitoring, characterized in that, The method comprises: acquiring current external wind data of a wind farm and current operating state data of a wind turbine generator in the wind farm, and generating a current cabin estimated disturbance range according to the current external wind data and the current operating state data; acquiring operating vibration detection data and operating noise detection data in a nacelle of the wind turbine generator, and generating a current operating estimated disturbance value according to the operating vibration detection data and the operating noise detection data, wherein the current operating estimated disturbance value is a value for indicating that the nacelle is affected by the operation of the wind turbine generator and external wind; determining whether the current operating estimated disturbance value is within the current cabin estimated disturbance range; if the determination is negative, generating a current operating early warning indication, and guiding a wind turbine generator maintenance personnel to perform maintenance on the nacelle of the wind turbine generator according to the current operating early warning indication; the nacelle comprises a plurality of vibration detection regions, each of the vibration detection regions is provided with a vibration detection point, one of the vibration detection points is provided with a vibration detection sensor, and the operating vibration detection data comprises a sum of data detected by activated vibration detection sensors, wherein one detection point operating vibration data is obtained after vibration detection by one activated vibration detection sensor; the nacelle further comprises a noise detection region, and the noise detection region is provided with a plurality of noise collection sensors for noise collection; acquiring operating vibration detection data and operating noise detection data in a nacelle of the wind turbine generator, and generating a current operating estimated disturbance value according to the operating vibration detection data and the operating noise detection data, comprises: acquiring a current wind level corresponding to the current external wind data, and generating a detection point activation number according to the current wind level; selecting to-be-activated detection points from the vibration detection points according to the detection point activation number; controlling vibration detection sensors corresponding to the to-be-activated detection points to start and acquiring detection point operating vibration data detected by the activated vibration detection sensors; generating an operating vibration interference value according to the detection point operating vibration data; calculating distances between noise collection sensors and to-be-activated detection points according to the to-be-activated detection points, and setting to-be-activated noise collection sensors, wherein the selected noise collection sensors are to-be-started noise sensors; starting the to-be-activated noise collection sensors and acquiring operating noise detection data; generating an operating noise interference value according to the operating noise detection data; adding the operating vibration interference value and the operating noise interference value to generate a current operating estimated disturbance value.

2. The wind farm intelligent early warning method based on operation state monitoring according to claim 1, characterized in that, generating an operating vibration interference value according to the detection point operating vibration data, comprises: calculating detection point fluctuation values according to the detection point operating vibration data, wherein one detection point fluctuation value corresponds to one detection point operating vibration data; adding the detection point fluctuation values to generate the operating vibration interference value.

3. The wind farm intelligent early warning method based on operation state monitoring according to claim 2, characterized in that, calculating one detection point fluctuation value according to one detection point operating vibration data, comprises: According to the detection point operation vibration data, a maximum vibration frequency, a minimum vibration frequency and an average vibration frequency are calculated; According to the detection point operation vibration data, a vibration frequency distribution curve is generated, and an ascending frequency segment and a descending frequency segment are screened from the vibration frequency distribution curve; The slopes of the ascending frequency segment and the descending frequency segment are calculated, and the absolute values of the slopes are taken as fluctuation slopes; A detection point fluctuation value is generated according to the maximum vibration frequency, the minimum vibration frequency, the average vibration frequency and the fluctuation slopes.

4. The wind farm intelligent early warning method based on operation state monitoring according to claim 3, characterized in that, According to the operation noise detection data, an operation noise interference value is generated, including: According to the operation noise detection data, a noise data frequency curve is generated; The total length of the noise data frequency curve is obtained, and an abnormal frequency curve is screened from the noise data frequency curve, wherein the abnormal frequency curve is a curve outside the normal frequency range in the noise data frequency curve; The abnormal curve length of the abnormal frequency curve is calculated; The ratio of the abnormal curve length to the total length of the curve is calculated, and an operation noise interference value is generated.

5. The wind farm intelligent early warning method based on operation state monitoring according to claim 1, characterized in that, Current external wind data of a wind farm and current operation state data of wind turbines in the wind farm are obtained, and a current cabin estimated disturbance range is generated according to the current external wind data and the current operation state data, including: The current external wind data of the wind farm is obtained, and the current wind force level is extracted from the current external wind data, and the current wind force disturbance coefficient is obtained according to the current wind force level; The used time of the cabin is obtained, and a disturbance amplification coefficient is generated according to the used time; The current operation state data of the wind turbines in the wind farm is obtained, and the current operation state data is compared with the preset conventional operation state data, and the conventional operation state data matched with the current operation state data is screened out; An initial disturbance range is obtained according to the conventional operation state data; The current cabin estimated disturbance range is generated according to the current wind force disturbance coefficient, the disturbance amplification coefficient and the initial disturbance range.

6. A wind farm intelligent early warning system based on operating state monitoring, characterized in that, The system includes: A state data analysis module is configured to obtain current external wind data of a wind farm and current operation state data of wind turbines in the wind farm, and generate a current cabin estimated disturbance range according to the current external wind data and the current operation state data; An operation disturbance generation module is configured to obtain operation vibration detection data and operation noise detection data in a cabin of the wind turbine, and generate a current operation estimated disturbance value according to the operation vibration detection data and the operation noise detection data, wherein the current operation estimated disturbance value is a value for indicating that the cabin is affected by the operation of the wind turbine and external wind force; An operation disturbance judgment module is configured to judge whether the current operation estimated disturbance value is within the current cabin estimated disturbance range; An operation early warning generation module is configured to generate a current operation early warning indication if the judgment is no, and guide wind turbine maintenance personnel to maintain the cabin of the wind turbine according to the current operation early warning indication. The machine cabin comprises a plurality of vibration detection areas, each of which is provided with a vibration detection point, and one of the vibration detection points is provided with a vibration detection sensor, the operation vibration detection data comprises the sum of the data detected by the activated vibration detection sensors, wherein one activated vibration detection sensor obtains one detection point operation vibration data after vibration detection; The machine cabin further comprises a noise detection area, and the noise detection area is provided with a plurality of noise collection sensors, and the noise collection sensors are used for noise collection; The operation disturbance generation module is further used for: obtaining a current wind level corresponding to the current external wind data, and generating a detection point activation number according to the current wind level; selecting to-be-activated detection points from the vibration detection points according to the detection point activation number; controlling the vibration detection sensors corresponding to the to-be-activated detection points to start and obtaining detection point operation vibration data detected by the activated vibration detection sensors; generating an operation vibration interference value according to the detection point operation vibration data; calculating the distance between the noise collection sensors and the to-be-activated detection points according to the to-be-activated detection points, and setting to-be-activated noise collection sensors, wherein the selected noise collection sensors are to-be-started noise sensors; starting the to-be-activated noise collection sensors and obtaining operation noise detection data; generating an operation noise interference value according to the operation noise detection data; adding the operation vibration interference value and the operation noise interference value to generate a current operation estimated disturbance value. 7.A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the computer device is configured to perform the method according to any one of claims 1-6 when the computer program is executed by the processor. The processor executes the computer program to realize the steps of the method in any one of claims 1 to 5.

8. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to realize the steps of the method in any one of claims 1 to 5.

Images