Lightning stroke risk assessment method, device and equipment for wind generating set in wind power plant

Abstract

The invention relates to a lightning stroke risk assessment method, device and equipment for a wind generating set in a wind power plant, in particular to the technical field of wind generating set monitoring, and the method comprises the steps: collecting multi-source set monitoring data corresponding to the wind generating set in real time through collection equipment configured for the wind generating set in the wind power plant; fusing multi-source unit monitoring data corresponding to the wind generating set, performing single-machine risk assessment on the wind generating set, and predicting a lightning stroke risk probability corresponding to the wind generating set; according to the lightning stroke risk probability corresponding to the wind generating set, determining a lightning stroke risk grade corresponding to the wind generating set; and finally, lightning stroke risk early warning of the lightning stroke risk level corresponding to the wind generating set is generated, and the lightning stroke risk early warning is used for carrying out risk response measure prompt corresponding to the lightning stroke risk level before the lightning stroke event occurs. The method can accurately reflect the actual risk of a single unit, and improves the operation safety and reliability of a wind power plant.

Description

Technical Field

[0001] This application relates to the field of wind turbine generator monitoring technology, and in particular to a method, device and equipment for assessing the lightning strike risk of wind turbine generators in a wind farm. Background Technology

[0002] Wind turbine generators (also known as generator sets) are usually installed in open areas or on mountaintops, making them susceptible to lightning strikes, which can lead to equipment damage, downtime for maintenance, and loss of power generation.

[0003] In related technologies, monitoring the location of lightning activity can provide a rough regional risk warning for wind farms, but it cannot predict the lightning risk of individual wind turbines in the wind farm, resulting in poor accuracy in lightning risk prediction. Summary of the Invention

[0004] In view of this, this application provides a method, device and equipment for assessing the lightning risk of wind turbine generators in a wind farm. The main purpose is to improve the technical problem in related technologies that, by monitoring the location of lightning activity, provide a rough regional risk warning for the wind farm, but cannot predict the lightning risk of a single unit in the wind farm, resulting in poor accuracy in lightning risk prediction.

[0005] The data acquisition equipment configured on each wind turbine in the wind farm collects multi-source unit monitoring data in real time. The multi-source unit monitoring data includes meteorological data, unit operation data, wind farm geographical data and wind farm lightning strike data for each wind turbine. By integrating monitoring data from multiple wind turbine generator sets, a single-unit risk assessment of the wind turbine generator set is conducted to predict the probability of lightning strike risk for the wind turbine generator set. The lightning risk level of the wind turbine generator set is determined based on the probability of lightning strike risk corresponding to the wind turbine generator set. Generate lightning risk warnings for the corresponding lightning risk levels of wind turbine generator sets. These warnings are used to provide guidance on risk response measures corresponding to the lightning risk level before a lightning strike occurs.

[0006] Secondly, this application provides a lightning strike risk assessment device for wind turbine generators in a wind farm, the device comprising: The data acquisition module is configured to collect real-time monitoring data from multiple wind turbine generators in the wind farm via data acquisition devices installed on each turbine. This multi-source monitoring data includes meteorological data, generator operation data, wind farm geographic data, and wind farm lightning strike data for each wind turbine generator. The prediction module is configured to integrate monitoring data from multiple wind turbine generator sets to perform individual risk assessments on the wind turbine generator sets and predict the probability of lightning strikes on the wind turbine generator sets. The determination module is configured to determine the lightning risk level of the wind turbine generator set based on the lightning risk probability corresponding to the wind turbine generator set. The generation module is configured to generate lightning risk warnings for the lightning risk level corresponding to the wind turbine generator set. The lightning risk warning is used to provide risk response measures corresponding to the lightning risk level before a lightning strike event occurs.

[0007] Thirdly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method of the first aspect.

[0008] Fourthly, this application provides an electronic device, including a storage medium, a processor, and a computer program stored on the storage medium and executable on the processor, wherein the processor executes the computer program to implement the method of the first aspect.

[0009] Fifthly, this application provides a computer program product, including a computer program that, when executed by a processor, implements the method of the first aspect.

[0010] By utilizing the above technical solution, this application provides a method, apparatus, and equipment for assessing the lightning strike risk of wind turbine generators in a wind farm. Compared with existing related technologies, this application uses data acquisition devices configured separately for each wind turbine generator in the wind farm to collect real-time monitoring data from multiple sources. This multi-source monitoring data includes meteorological data, generator operation data, wind farm geographical data, and wind farm lightning strike data for each wind turbine generator. Then, the multi-source monitoring data is integrated to perform a single-unit risk assessment of the wind turbine generator, predicting the probability of lightning strike risk for each wind turbine generator. Based on the probability of lightning strike risk, the lightning strike risk level of each wind turbine generator is determined. Finally, a lightning strike risk warning is generated for the corresponding lightning strike risk level, which serves as a reminder of risk response measures before a lightning strike event occurs. This application can collect multi-source unit monitoring data corresponding to each wind turbine in a wind farm in real time. It comprehensively considers multi-source data such as meteorological data, unit operation data, wind farm geographical data, and wind farm lightning strike data. Then, it integrates the multi-source unit monitoring data corresponding to each wind turbine to conduct a single-unit risk assessment for each wind turbine, generate the lightning strike risk probability for each wind turbine, and generate real-time early warning based on the lightning strike risk probability. According to the lightning strike risk level, it generates a lightning strike risk warning before a lightning strike event occurs, constructs an active protection mechanism to cope with sudden weather changes, accurately reflects the actual risk of a single unit, and thus improves the operational safety and reliability of the wind farm.

[0011] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the following are specific embodiments of this application. Attached Figure Description

[0012] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0013] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0014] Figure 1 A flowchart illustrating a lightning strike risk assessment method for wind turbine generators in a wind farm, provided in an embodiment of this application, is shown. Figure 2This paper presents a schematic diagram of a lightning strike risk assessment device for wind turbine generators in a wind farm, as provided in an embodiment of this application. Detailed Implementation

[0015] To better understand the above-mentioned objectives, features, and advantages of this application, the solution of this application will be further described below. It should be noted that, unless otherwise specified, the embodiments and features described in these embodiments can be combined with each other.

[0016] To address the issue that existing technologies, which rely on monitoring lightning activity locations to provide coarse regional risk assessments of wind farms but fail to predict the lightning strike risk of individual wind turbines, resulting in poor accuracy in lightning risk prediction, this embodiment provides a method for assessing the lightning strike risk of wind turbines in a wind farm. Figure 1 As shown, the method includes: Step 101: Collect real-time monitoring data of the multi-source units corresponding to the wind turbine generators in the wind farm using the data acquisition equipment configured on each wind turbine generator.

[0017] The multi-source unit monitoring data includes meteorological data, unit operation data, wind field geographical data, and wind field lightning strike data for each wind turbine.

[0018] In some embodiments, corresponding data acquisition devices can be installed at designated locations for each wind turbine or wind farm to sense and collect various monitoring parameters of each wind turbine, such as meteorological data acquisition devices, unit status data acquisition devices, lightning strike monitoring data acquisition devices, and geographic location data acquisition devices. These devices convert signals such as the environment and equipment status of the wind turbine into transmittable and analyzable digital data for multi-source data fusion processing.

[0019] Correspondingly, multi-source unit monitoring data refers to datasets acquired through different types of acquisition equipment that cover the full-dimensional operation and environmental status of wind turbine generators. This allows for the acquisition of diverse fusion data sources, enabling comprehensive analysis and prediction based on multiple dimensions, thereby improving the accuracy of lightning strike risk prediction for each wind turbine generator. Specifically, meteorological data reflects the atmospheric environment of the area where the wind turbine is located, serving as fundamental data for assessing wind resources and predicting extreme weather. This includes temperature, humidity, air pressure, wind speed, wind direction, and atmospheric electric field strength. Unit operation data directly characterizes the core parameters of the wind turbine's own operation, used to determine whether the unit is generating electricity normally and whether there are potential faults. This includes rotor speed, generator power, grounding status, gearbox temperature, and pitch angle. Wind farm geographic data describes the spatial location and topographic features of the wind turbine and wind farm, used to analyze the impact of the geographical environment on unit operation and lightning strike risk. This includes the unit's latitude and longitude, altitude, terrain type (e.g., highlands, lowlands, ridges, valleys, plains), and the distribution of surrounding obstacles. Wind farm lightning strike data records relevant parameters of the wind farm and individual units subjected to lightning strikes. This is crucial data for assessing lightning strike risk and analyzing lightning faults, including peak lightning current, lightning polarity, lightning strike time, lightning strike location, and changes in electric field strength during the strike.

[0020] Step 102: Integrate the monitoring data of the multi-source units corresponding to the wind turbine generator set, conduct a single-unit risk assessment of the wind turbine generator set, and predict the probability of lightning strike risk corresponding to the wind turbine generator set.

[0021] In some embodiments, the lightning strike risk probability can be a quantified value of the predicted likelihood of a single wind turbine being struck by lightning. Specifically, monitoring data from different sources and dimensions, such as meteorological data, turbine operation data, wind farm geographic data, and wind farm lightning strike data, can be integrated, calibrated, and correlated using specific algorithms (such as weighted fusion, machine learning fusion, and spatiotemporal correlation fusion). Based on multi-source data, risk prediction can be performed on a single wind turbine, focusing on factors such as the turbine's own operating status, environment, and historical lightning strikes. This quantitatively assesses the turbine's lightning strike risk. By using a wind turbine lightning strike risk assessment method based on multi-source data fusion and machine learning algorithms, the limitations of a single data source can be eliminated, data complementarity and synergy can be achieved, and real-time, accurate, and single-unit-level lightning strike risk early warning can be realized, improving the safety and reliability of wind farm operation and increasing the accuracy of assessment.

[0022] Step 103: Determine the lightning risk level of the wind turbine generator set based on the lightning strike risk probability corresponding to the wind turbine generator set.

[0023] In specific application scenarios, probability threshold ranges for lightning strike risk corresponding to different lightning strike risk levels can be set to classify and categorize the lightning strike risk of wind turbine generators. The division of probability threshold ranges can be preset based on the actual operation and maintenance needs of the wind farm and lightning protection specifications. For example, it can be divided into multiple risk levels (such as extremely low, low, medium, high, and extremely high), with different levels corresponding to differentiated protection strategies and early warning measures.

[0024] Step 104: Generate a lightning risk warning for the lightning risk level corresponding to the wind turbine generator set.

[0025] Among them, the lightning strike risk warning is used to provide risk response measures corresponding to the lightning strike risk level before a lightning strike event occurs.

[0026] In some embodiments, the output methods for lightning risk warnings may include: audible and visual alarms, mobile app push notifications, and linkage with automated control systems. A preset time range for the warning's activation (e.g., 2 hours or 6 hours in the future) can also be used to facilitate maintenance personnel in scheduling response times. The warning can be lifted once the specified time range is reached, or when the probability of a lightning strike is detected to have decreased to a preset threshold. Correspondingly, maintenance personnel can inspect the unit, such as checking for lightning strike marks on the blades, verifying the grounding resistance, checking the surge arrester's status, and assessing whether the unit can resume normal power generation. Simultaneously, a warning handling log is generated, recording the warning time, implemented measures, equipment status, and other information to optimize subsequent lightning risk warnings.

[0027] For example, a lightning strike risk warning can be displayed via a pop-up window on the large screen of the Geographic Information System (GIS) in the central control room, accompanied by audible and visual alarms; lightning strike risk warning information, along with a risk level indicator, can also be pushed to the terminal APP of maintenance personnel. Risk response measures may include: blade angle adjustment, active power adjustment, and shutdown.

[0028] By applying the technical solution of this application embodiment, this embodiment can collect multi-source unit monitoring data corresponding to each wind turbine in the wind farm in real time. It comprehensively considers multi-source data such as meteorological data, unit operation data, wind farm geographical data, and wind farm lightning strike data. Then, it integrates the multi-source unit monitoring data corresponding to each wind turbine to conduct a single-unit risk assessment for each wind turbine, generate the lightning strike risk probability corresponding to each wind turbine, and conduct real-time early warning based on the lightning strike risk probability. According to the lightning strike risk level, it generates a lightning strike risk warning before the occurrence of a lightning strike event, builds an active protection mechanism to cope with sudden weather changes, accurately reflects the actual risk of a single unit, and thus improves the safety and reliability of wind farm operation.

[0029] Furthermore, as a refinement and extension of the specific implementation of the above embodiments, in order to fully illustrate the implementation of this embodiment, the data acquisition device includes a meteorological sensor, a unit status sensor, and a lightning detection device. Optionally, step 101 may specifically include: acquiring meteorological data corresponding to the wind turbine generator through a meteorological sensor configured separately for the wind turbine generator; acquiring unit operation data corresponding to the wind turbine generator through a unit status sensor configured separately for the wind turbine generator; acquiring wind farm lightning strike data corresponding to the wind turbine generator through a lightning detection device configured separately for the wind turbine generator; and acquiring wind farm geographic data corresponding to the wind turbine generator through a geographic information system corresponding to the wind farm, wherein the geographic information system is used to input terrain data of the wind turbine generators in the wind farm.

[0030] Among them, the separately configured meteorological sensor can refer to the sensing device that is independently deployed for a single wind turbine generator set to sense the micro-meteorological environment around it. Unlike the meteorological stations that are centrally deployed in the wind farm, it can accurately capture the real-time meteorological parameters of the generator set's location and avoid the influence of spatial differences in regional meteorological data. It may include, but is not limited to, temperature sensors, humidity sensors, air pressure sensors, atmospheric electric field strength sensors, etc. A separately configured unit status sensor can refer to a sensing device installed on a key component of a wind turbine generator set to monitor the unit's own operating status. It is directly related to the unit's power generation performance and equipment health. It can be linked with the unit's control system to provide real-time feedback on the unit's operating parameters. These sensors may include, but are not limited to, speed sensors, voltage / current sensors, grounding status sensors, and temperature sensors. A separately configured lightning detection device can refer to a dedicated lightning monitoring device deployed to meet the lightning protection needs of a single wind turbine generator set. It is used to capture the electrical parameters and location information of the generator set when it is struck by lightning, so as to achieve accurate recording and tracing of single-unit lightning strike events. The lightning detection device may include, but is not limited to, lightning current sensors, electric field change monitoring modules, etc. The geographic information system corresponding to the wind farm can serve as a spatial information management platform at the wind farm level. By integrating preliminary surveying data and real-time location information, it can input and manage the topographic data and spatial relationships of all wind turbine generators, providing geographic dimension support for single-unit risk assessment. Among them, the topographic data in the wind farm geographic data can refer to the topographic and geomorphological feature parameters of the location of the wind turbine generator, including but not limited to elevation, terrain type, slope, and distribution of surrounding obstacles, which are used to analyze the impact of the geographical environment on lightning strike risk.

[0031] For example, an integrated meteorological sensor (such as a PT1000 temperature sensor, capacitive humidity sensor, piezoresistive barometric pressure sensor, and atmospheric electric field meter) can be installed on the top of the unit; an ultrasonic anemometer can be installed at the bottom of the unit tower; a Hall effect speed sensor can be installed at the rotor shaft end; a Hall voltage / current sensor can be installed in the generator grid-connection cabinet; a grounding resistance sensor can be installed in the tower grounding circuit; and a thermistor temperature sensor can be installed in the gearbox oil circuit; a Rogowski coil lightning current sensor can be installed at the root of the unit blades; a lightning positioning substation can be installed on the top of the tower; and an electric field mutation monitoring module can be added to the outer wall of the nacelle; a Global Navigation Satellite System (GNSS) positioning module can be installed on the unit's basic platform; and topographic mapping can be completed on the wind farm side using a UAV equipped with a Light Detection and Ranging (LiDAR) system, with the data accessed to the wind farm GIS platform.

[0032] Correspondingly, during unit operation, the meteorological sensors on the top of the nacelle can be used to perceive ambient temperature, humidity, air pressure, and atmospheric electric field strength in real time, and data sampling can be completed every 10 seconds; the ultrasonic anemometer can collect wind speed and direction data every 5 seconds to obtain meteorological data, and then transmit the meteorological data to the unit's main control system via RS485 bus.

[0033] Data Acquisition Process: The rotor rotation frequency can be monitored in real time by a Hall effect speed sensor and converted into rotor speed; voltage / current sensors in the grid-connected cabinet collect phase voltage and phase current, and the main control system calculates active power using formulas; the grounding resistance sensor automatically detects the grounding loop resistance once per hour to obtain unit operating data, which can be uploaded to the Supervisory Control and Data Acquisition (SCADA) system at a frequency of 2 seconds per upload; the precise latitude, longitude, and altitude of different numbered units (such as WTG-08) can be obtained through the GNSS positioning module; and the terrain data mapped by UAV LiDAR is imported into the GIS platform to generate a wind farm elevation model, which, after being associated with the unit location, forms a geographic information layer as wind farm geographic data. The wind farm geographic data can be updated quarterly (or annually if the terrain remains unchanged).

[0034] When a lightning strike occurs, the lightning current waveform can be captured by the Rogowski coil sensor at the blade root, recording the peak value and polarity; the lightning location substation can synchronize the lightning strike time and location information to the wind farm lightning monitoring platform; the electric field mutation module records the change in electric field intensity before and after the lightning strike to obtain the corresponding wind farm lightning strike data for the unit. The wind farm lightning strike data is uploaded in a triggered manner, and it goes into hibernation when there is no lightning strike, and uploads immediately when there is a lightning strike.

[0035] In specific application scenarios, the multi-source unit monitoring data of each unit can be uploaded to the wind farm intelligent operation and maintenance platform corresponding to the wind farm. The platform integrates atmospheric electric field intensity, terrain data and historical lightning strike records to determine the lightning risk level of WTG-08 before a thunderstorm arrives, so as to carry out lightning risk warning, such as triggering blade feathering operation in advance to reduce the probability of lightning damage. Through this multi-source data fusion technology, meteorological, unit and geographical multi-dimensional data can be efficiently integrated to build a multi-source data fusion lightning risk assessment architecture, forming a single-unit level real-time early warning mechanism and system integration solution.

[0036] Optionally, step 102 may specifically include: preprocessing the multi-source unit monitoring data; inputting the preprocessed multi-source unit monitoring data into a preset lightning strike risk prediction model to perform a single-unit risk assessment of the wind turbine generator, obtaining the lightning strike risk probability corresponding to the wind turbine generator, the preset lightning strike risk prediction model being used to extract the lightning strike risk features corresponding to the multi-source unit monitoring data, and fusing the lightning strike risk features to obtain the lightning strike risk probability corresponding to the wind turbine generator.

[0037] Among them, the preset lightning strike risk prediction model can be a machine learning model, a deep learning model, a support vector machine, a gradient boosting tree, etc.; the preset lightning strike risk prediction model can be built and trained using multi-source unit monitoring data collected in history by the unit, such as supervised learning based on historical lightning strike event data, which is suitable for lightning strike risk feature extraction and selection in wind farms, ensuring system response speed and reliability, and having real-time data processing and model inference capabilities.

[0038] For example, a deep learning-based lightning risk prediction system can be constructed, using convolutional neural networks to process spatial meteorological data and recurrent neural networks to process time series data. The workflow may include: inputting multi-source unit monitoring data into a preset lightning risk prediction model, performing fusion processing on the preset lightning risk prediction model, and outputting a risk probability map based on the model. This system is suitable for scenarios with large data volumes and high-dimensional features, thereby improving the accuracy of lightning risk prediction.

[0039] Optionally, preprocessing of multi-source turbine monitoring data may include: normalizing temperature, humidity, and air pressure in meteorological data corresponding to wind turbine generators; statistically analyzing the frequency of anomalies in electric field intensity data according to the time dimension; filtering out abnormal fluctuations in speed and power in turbine status data corresponding to wind turbine generators; performing spatiotemporal calibration on peak lightning current and lightning strike time in lightning detection data corresponding to wind turbine generators and matching them to the geographical coordinates of the corresponding wind turbine generators; extracting the terrain type and relative elevation to surrounding high points from the wind field geographical data corresponding to the wind turbine generators; and obtaining the terrain lightning susceptibility coefficient corresponding to the wind turbine generators based on the terrain type and relative elevation to surrounding high points.

[0040] For example, before training a preset lightning strike risk prediction model using historically collected multi-source unit monitoring data, or before inputting real-time collected multi-source unit monitoring data into the preset lightning strike risk prediction model for prediction, the historically collected multi-source unit monitoring data can be preprocessed, such as normalization, anomaly filtering, spatiotemporal calibration, and feature extraction. This preprocessing can eliminate data differences, correct biases, and extract key features, providing high-quality data input for subsequent model training, fusion, and risk calculation, and ensuring the accuracy and reliability of subsequent analysis results.

[0041] Optionally, step 102 may further include: extracting lightning risk features corresponding to the monitoring data of the multi-source units, including meteorological features, unit features, geographical features, and lightning features; and fusing the lightning risk features corresponding to each wind turbine unit to generate the lightning risk probability corresponding to each wind turbine unit.

[0042] For example, meteorological characteristics may include, but are not limited to, the number of thunderstorm days, the frequency of atmospheric electric field anomalies, the intensity of atmospheric electric field, and the movement speed of thunderstorm clouds; unit characteristics may include the grounding resistance of the single-unit grounding status sensor, the aging / fault records of the blade lightning protection device, the conductivity of the blade down conductor, and the number of past lightning strikes to the unit; geographical characteristics may include the unit's terrain elevation, the degree of overlap with the movement path of thunderclouds, and spatial data on the distribution of surrounding lightning protection facilities (such as high-voltage lines and mountains); lightning strike characteristics may include the lightning strike density and peak lightning current in the wind field area. In this way, characteristic variables strongly correlated with lightning strike risk are extracted and quantified.

[0043] Optionally, atmospheric electric field intensity from meteorological data can be converted into an electric field risk coefficient (e.g., 32kV / m corresponds to a coefficient of 0.85, with a threshold >20kV / m indicating a high-risk range); thunderstorm cloud movement speed can be converted into a time warning coefficient (e.g., 15km / h corresponds to a coefficient of 0.7, representing that the thunderstorm cloud will arrive within 1 hour); grounding resistance from unit operation data can be converted into a lightning protection shortcoming coefficient (e.g., 5.8Ω corresponds to a coefficient of 0.6, with >4Ω being a critical value); blade down conductor conductivity can be converted into a protection failure coefficient (e.g., 92% corresponds to a coefficient of 0.4, with <90% indicating a high risk); ridge topography from geographical data can be converted into a geographical susceptibility coefficient (e.g., ridge corresponds to a coefficient of 1.0, plains to 0.5); and regional lightning strike density from lightning strike data can be converted into a historical risk coefficient (6.2 strikes / km²). Annual coefficient of 0.9, >5 times / km² (The annual density is extremely high).

[0044] Further optional, the weights of the coefficients of each dimension corresponding to each wind turbine can be weighted and integrated (e.g., 30% weight for the coefficient corresponding to meteorological data, 25% weight for the coefficient corresponding to the turbine status data, 20% weight for the coefficient corresponding to geographical data, and 25% weight for the coefficient corresponding to lightning strike data). Based on the coefficient weights corresponding to different data, the lightning strike risk characteristics of each wind turbine can be integrated to generate the lightning strike risk probability of each wind turbine.

[0045] Optionally, the method in this embodiment further includes: generating a risk probability map corresponding to the wind turbine generator set based on the lightning strike risk probability corresponding to the wind turbine generator set. The risk probability map includes risk probability maps corresponding to the wind turbine generator set and risk probability maps corresponding to the wind farm.

[0046] The risk probability map can take the form of a heat map, a trend line chart, or a bar chart. For example, a spatial heat map of the wind farm can be generated using ArcGIS based on the lightning strike risk probability of each unit, and multi-dimensional data visualization can be achieved using Origin. Alternatively, a bar or line chart can be drawn using Excel based on the lightning strike risk probability of each unit.

[0047] Optionally, step 103 may specifically include: determining the lightning risk level corresponding to each wind turbine generator based on the lightning risk probability and multi-level risk thresholds displayed on the risk probability map corresponding to the wind turbine generator.

[0048] For example, a risk probability map can be a heat map, with the horizontal axis representing the number of wind turbines and the vertical axis representing the probability level of a single turbine being struck by lightning. Multiple risk thresholds corresponding to different risk levels can be configured to classify the risk based on preset lightning strike probabilities, using different colors to represent different lightning strike risk levels, with darker colors indicating a higher probability of a turbine being struck by lightning. By integrating meteorological data, turbine operating status, topography, and historical lightning strike data, a machine learning model can be constructed to achieve dynamic assessment and graded early warning of lightning strike risk for wind turbines.

[0049] As one possible implementation, a corresponding system can be constructed based on the method of this embodiment, and the system components may include: Data acquisition layer: includes meteorological sensors (such as temperature sensors, humidity sensors, air pressure sensors, electric field strength sensors), unit status sensors (speed, power, grounding status), lightning detection equipment, and geographic information system; Data processing layer: Data preprocessing module, which performs preprocessing operations such as outlier handling, data alignment, and feature extraction on multi-source unit monitoring data; Risk assessment model: Construct a pre-defined lightning strike risk prediction model based on random forest or neural network. Input features include real-time meteorological data, unit operation data, terrain elevation, historical lightning strike frequency, etc. Early warning output layer: Outputs early warning signals based on the lightning risk level predicted by the preset lightning risk prediction model, and supports visualization and automatic push.

[0050] The specific workflow corresponding to this system may include: (1) Monitoring data from multi-source units are collected in real time and uploaded to the central processing unit; (2) Extraction of lightning strike risk characteristics and fusion of monitoring data from multiple generator units; (3) Input the preset lightning risk prediction model to conduct lightning risk assessment and output the lightning risk level of each unit (such as low risk, medium risk, high risk). (4) If the lightning strike risk level reaches the preset warning threshold, the warning mechanism will be triggered to notify the operation and maintenance personnel or start the protection system.

[0051] Compared with related technologies, the embodiments of this application can collect multi-source unit monitoring data through different acquisition devices configured separately for each unit, extract lightning strike risk characteristics from different data dimensions, fuse the lightning strike risk characteristics of each unit to obtain the lightning strike risk probability of each unit, and visualize and warn through risk probability maps. This comprehensive consideration of multi-source data can improve the assessment accuracy by more than 50%, achieving high-precision assessment, and the early warning response time can be less than 1 minute. It supports active protection, realizes real-time early warning, and performs independent risk assessment for each unit, achieving more targeted single-unit-level assessment. Furthermore, the preset lightning strike risk prediction model can be continuously optimized with data accumulation, adapt to different wind field environments, and has adaptive learning capabilities. At the same time, this early warning method effectively reduces downtime losses and equipment maintenance costs caused by lightning strikes, thereby reducing operation and maintenance costs.

[0052] Furthermore, embodiments of this application provide a lightning strike risk assessment device for wind turbine generators in a wind farm, such as... Figure 2 As shown, the device includes: a data acquisition module 31, a prediction module 32, a determination module 33, and a generation module 34.

[0053] The acquisition module 31 is configured to acquire real-time multi-source unit monitoring data corresponding to the wind turbine generators through the acquisition devices configured on the wind turbine generators in the wind farm. The multi-source unit monitoring data includes meteorological data, unit operation data, wind farm geographical data and wind farm lightning strike data corresponding to each wind turbine generator. Prediction module 32 is configured to integrate monitoring data of multiple wind turbine generator sets to perform single-unit risk assessment of wind turbine generator sets and predict the probability of lightning strike risk corresponding to wind turbine generator sets. Module 33 is configured to determine the lightning risk level of the wind turbine generator set based on the lightning risk probability corresponding to the wind turbine generator set. The generation module 34 is configured to generate a lightning risk warning for the lightning risk level corresponding to the wind turbine generator set. The lightning risk warning is used to provide a prompt for risk response measures corresponding to the lightning risk level before a lightning event occurs.

[0054] In some embodiments, the prediction module 32 is specifically configured to preprocess the multi-source unit monitoring data; input the preprocessed multi-source unit monitoring data into a preset lightning strike risk prediction model to perform a single-unit risk assessment of the wind turbine generator, obtain the lightning strike risk probability corresponding to the wind turbine generator, and use the preset lightning strike risk prediction model to extract the lightning strike risk features corresponding to the multi-source unit monitoring data, and fuse the lightning strike risk features to obtain the lightning strike risk probability corresponding to the wind turbine generator.

[0055] In some embodiments, the prediction module 32 is specifically configured to: normalize the temperature, humidity, and air pressure in the meteorological data corresponding to the wind turbine generator set; statistically analyze the abnormal frequency of the electric field intensity data according to the time dimension; filter abnormal fluctuation values ​​of speed and power in the unit status data corresponding to the wind turbine generator set; perform spatiotemporal calibration on the lightning current peak value and lightning strike time in the lightning detection data corresponding to the wind turbine generator set, and match them to the geographical coordinates of the corresponding wind turbine generator set; extract the terrain type and relative elevation with the surrounding high points from the wind field geographical data corresponding to the wind turbine generator set; and obtain the terrain lightning susceptibility coefficient corresponding to the wind turbine generator set based on the terrain type and relative elevation with the surrounding high points.

[0056] In some embodiments, the generation module 34 is specifically configured to generate a risk probability map corresponding to the wind turbine generator set based on the lightning strike risk probability corresponding to the wind turbine generator set. The risk probability map includes a risk probability map corresponding to the wind turbine generator set and a risk probability map corresponding to the wind farm.

[0057] In some embodiments, the prediction module 32 is specifically configured to determine the lightning risk level corresponding to each wind turbine generator based on the lightning risk probability and multi-level risk thresholds displayed in the risk probability map corresponding to the wind turbine generator.

[0058] In some embodiments, the prediction module 32 is further configured to extract lightning risk features corresponding to multi-source unit monitoring data, including meteorological features, unit features, geographical features, and lightning features; and to fuse the lightning risk features corresponding to each wind turbine unit to generate the lightning risk probability corresponding to each wind turbine unit.

[0059] In some embodiments, the data acquisition device includes a meteorological sensor, a turbine status sensor, and a lightning detection device; the acquisition module 31 is specifically configured to acquire meteorological data corresponding to the wind turbine generator through the meteorological sensor configured separately for the wind turbine generator; acquire turbine operation data corresponding to the wind turbine generator through the turbine status sensor configured separately for the wind turbine generator; acquire wind farm lightning strike data corresponding to the wind turbine generator through the lightning detection device configured separately for the wind turbine generator; and acquire wind farm geographic data corresponding to the wind turbine generator through the geographic information system corresponding to the wind farm, wherein the geographic information system is used to input the terrain data of the wind turbine generators in the wind farm.

[0060] It should be noted that other corresponding descriptions of the functional units involved in the lightning strike risk assessment device for wind turbine generators in a wind farm provided in this application embodiment can be found by referring to... Figure 1 The corresponding description in [the document] will not be repeated here.

[0061] Based on the above, Figure 1As illustrated in the example, correspondingly, embodiments of this application also provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-described... Figure 1 The example method shown.

[0062] Based on the above, Figure 1 As illustrated, correspondingly, embodiments of this application also provide a computer program product, including a computer program that, when executed by a processor, implements the above-described... Figure 1 The example method shown.

[0063] Based on this understanding, the technical solutions of the embodiments of this application can be embodied in the form of a software product. The software product can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, mobile hard drive, etc.) and includes several instructions to cause a computer device (such as a personal computer, server, or network device, etc.) to execute the methods of various implementation scenarios of this application.

[0064] Based on the above, Figure 1 The method shown, and Figure 2 To achieve the above objectives, the present application also provides an electronic device, comprising a storage medium and a processor; the storage medium for storing a computer program; and the processor for executing the computer program to implement the above-described virtual device embodiments. Figure 1 The method shown.

[0065] Optionally, the aforementioned electronic device may also include a user interface, a network interface, a camera, radio frequency (RF) circuitry, sensors, audio circuitry, a Wi-Fi module, etc. The user interface may include a display screen, an input unit, etc.

[0066] Those skilled in the art will understand that the physical device structure provided in this embodiment does not constitute a limitation on the physical device, and may include more or fewer components, or combine certain components, or have different component arrangements.

[0067] The storage medium may also include an operating system and a network communication module. The operating system is a program that manages the hardware and software resources of the aforementioned physical device, supporting the operation of information processing programs and other software and / or programs. The network communication module is used to enable communication between the various components within the storage medium, as well as communication with other hardware and software in the information processing physical device.

[0068] Through the above description of the embodiments, those skilled in the art can clearly understand that this application can be implemented using software plus necessary general-purpose hardware platforms, or it can be implemented through hardware. The embodiments of this application can collect multi-source unit monitoring data through different acquisition devices configured separately for each unit, extract lightning strike risk characteristics from different data dimensions, fuse the lightning strike risk characteristics of each unit to obtain the lightning strike risk probability of each unit, and visualize and warn through a risk probability map. This comprehensive consideration of multi-source data can improve the assessment accuracy by more than 50%, achieving high-precision assessment. The early warning response time can be less than 1 minute, supporting active protection and achieving real-time early warning. Independent risk assessment is performed for each unit, achieving more targeted single-unit-level assessment. Furthermore, the preset lightning strike risk prediction model can be continuously optimized with data accumulation, adapting to different wind field environments and possessing adaptive learning capabilities. This early warning method effectively reduces downtime losses and equipment maintenance costs caused by lightning strikes, thereby reducing operation and maintenance costs.

[0069] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the term "comprising" or any other variations thereof is intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes the element.

[0070] The above are merely specific embodiments of this application, enabling those skilled in the art to understand or implement this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to these embodiments, but is to be accorded the widest scope consistent with the principles and novel features claimed herein.

Claims

1. A method for assessing the lightning strike risk of wind turbine generators in a wind farm, characterized in that, include: The data acquisition equipment configured on each wind turbine in the wind farm collects multi-source unit monitoring data in real time. The multi-source unit monitoring data includes meteorological data, unit operation data, wind farm geographical data, and wind farm lightning strike data for each wind turbine. By integrating the monitoring data of the multi-source units corresponding to the wind turbine generator set, a single-unit risk assessment is performed on the wind turbine generator set to predict the probability of lightning strike risk corresponding to the wind turbine generator set. The lightning strike risk level of the wind turbine generator set is determined based on the lightning strike risk probability corresponding to the wind turbine generator set. A lightning risk warning is generated for the lightning risk level corresponding to the wind turbine generator set. The lightning risk warning is used to provide a prompt for risk response measures corresponding to the lightning risk level before a lightning strike event occurs.

2. The method according to claim 1, characterized in that, The process of integrating multi-source turbine monitoring data corresponding to the wind turbine to perform a single-unit risk assessment of the wind turbine and predicting the probability of lightning strike risk corresponding to the wind turbine includes: The monitoring data of the multi-source units are preprocessed; The preprocessed multi-source unit monitoring data is input into a preset lightning strike risk prediction model to perform a single-unit risk assessment on the wind turbine generator set and obtain the lightning strike risk probability corresponding to the wind turbine generator set. The preset lightning strike risk prediction model is used to extract the lightning strike risk features corresponding to the multi-source unit monitoring data and fuse the lightning strike risk features to obtain the lightning strike risk probability corresponding to the wind turbine generator set.

3. The method according to claim 2, characterized in that, The preprocessing of the monitoring data from the multi-source generating units includes: The temperature, humidity, and air pressure in the meteorological data corresponding to the wind turbine generator are normalized, and the frequency of anomalies in the electric field intensity data is statistically analyzed according to the time dimension. Filter out abnormal fluctuations in speed and power in the unit status data corresponding to the wind turbine generator set; The peak lightning current and lightning strike time in the lightning detection data corresponding to the wind turbine generator are spatiotemporally calibrated and matched to the geographical coordinates of the corresponding wind turbine generator. Based on the wind field geographic data corresponding to the wind turbine generator set, the terrain type where the wind turbine generator set is located and its relative elevation with the surrounding high points are extracted. Based on the terrain type and the relative elevation with the surrounding high points, the terrain lightning susceptibility coefficient corresponding to the wind turbine generator set is obtained.

4. The method according to claim 2, characterized in that, The method further includes: Based on the lightning strike risk probability corresponding to the wind turbine generator set, a risk probability map corresponding to the wind turbine generator set is generated. The risk probability map includes risk probability maps corresponding to the wind turbine generator set and risk probability maps corresponding to the wind farm.

5. The method according to claim 4, characterized in that, The step of determining the lightning strike risk level corresponding to the wind turbine generator set based on the lightning strike risk probability corresponding to the wind turbine generator set includes: Based on the lightning strike risk probability and multi-level risk thresholds displayed on the risk probability map corresponding to the wind turbine generator set, the lightning strike risk level corresponding to the wind turbine generator set is determined.

6. The method according to claim 1, characterized in that, The method of integrating multi-source unit monitoring data corresponding to the wind turbine generator set to perform a single-unit risk assessment of the wind turbine generator set and predicting the probability of lightning strike risk corresponding to the wind turbine generator set also includes: Extract the lightning risk characteristics corresponding to the multi-source unit monitoring data, wherein the lightning risk characteristics include meteorological characteristics, unit characteristics, geographical characteristics, and lightning characteristics; The lightning strike risk characteristics of each wind turbine are integrated to generate the lightning strike risk probability for each wind turbine.

7. The method according to claim 1, characterized in that, The data acquisition equipment includes meteorological sensors, unit status sensors, and lightning detection equipment; The data acquisition equipment configured on each wind turbine in the wind farm collects real-time monitoring data of the multi-source units corresponding to the wind turbines, including: Meteorological data corresponding to the wind turbine generator is obtained through the meteorological sensors configured separately for the wind turbine generator. The unit operation data corresponding to the wind turbine generator is obtained through the unit status sensor configured separately for the wind turbine generator; The lightning strike data of the wind farm corresponding to the wind turbine is obtained through the lightning detection equipment configured separately for the wind turbine. Geographic data of the wind farm corresponding to the wind turbine generator is obtained through the geographic information system corresponding to the wind farm. The geographic information system is used to input the terrain data of the wind turbine generator in the wind farm.

8. A lightning strike risk assessment device for wind turbine generators in a wind farm, characterized in that, include: The data acquisition module is configured to collect real-time monitoring data of the multi-source units corresponding to the wind turbine generators through the data acquisition devices configured on each wind turbine generator in the wind farm. The multi-source unit monitoring data includes meteorological data, unit operation data, wind farm geographical data and wind farm lightning strike data corresponding to each wind turbine generator. The prediction module is configured to integrate multi-source unit monitoring data corresponding to the wind turbine generator set, perform single-unit risk assessment of the wind turbine generator set, and predict the probability of lightning strike risk corresponding to the wind turbine generator set. The determination module is configured to determine the lightning risk level corresponding to the wind turbine generator set based on the lightning risk probability corresponding to the wind turbine generator set; The generation module is configured to generate a lightning risk warning for the lightning risk level corresponding to the wind turbine generator set. The lightning risk warning is used to provide a prompt for risk response measures corresponding to the lightning risk level before a lightning strike event occurs.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method of any one of claims 1 to 7.

10. An electronic device comprising a storage medium, a processor, and a computer program stored on the storage medium and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method of any one of claims 1 to 7.

Images