Mountainous ground distributed wind power grounding and lightning protection system and construction method

Abstract

The present application relates to the technical field of wind power lightning protection, in particular to a mountainous distributed wind power grounding and lightning protection system and a construction method, the present application generates a dynamic mountainous dispersion characteristic map with temperature and moisture content slope correction and terrain lightning attraction correction through a mountainous grounding feature dynamic modeling unit, realizes multi-factor constraint dispersion weight dynamic distribution and thunderstorm period cross-point capacity transfer through a grounding point dispersion collaborative optimization unit, completes terrain correction thunderstorm early warning and frequency related attenuation surge calculation to trigger the discharge passage in advance through a cable transition lightning protection active control unit, realizes operation state feedback and model closed-loop iteration through a multi-node state fusion management and control unit, which can improve the dispersion utilization rate and lightning protection reliability of mountainous scenes, and is suitable for mountainous distributed wind power scenes.

Description

Technical Field

[0001] This invention relates to the field of lightning protection technology for wind power generation, and more specifically, to a distributed wind power grounding and lightning protection system and construction method for mountainous areas. Background Technology

[0002] Existing grounding and lightning protection technologies, after decades of engineering practice and iteration, have formed a complete standard system and mature implementation plans. In scenarios such as centralized wind farms in plains areas, urban substations, and industrial plants, where soil parameters are relatively uniform, grounding grids are centrally and continuously arranged, and thunderstorm activity patterns are stable, stable and reliable lightning protection effects can be achieved. The relevant parameter thresholds and control logic are designed around the typical characteristics of the above-mentioned conventional scenarios, which can meet the protection requirements of conventional scenarios. As wind power development extends to mountainous resource areas, the turbine sites of distributed wind power projects in mountainous areas are mostly distributed along ridges. The terrain in the project area has a large undulation, and there are significant differences in soil type and thickness at different locations. The soil conductivity dynamically changes with local temperature, moisture content, and slope runoff. The lightning attraction effect of prominent terrain such as ridges makes thunderstorm activity more pronounced. Due to the local distribution characteristics of mountainous distributed wind power, the collection lines are mostly laid in a mixed manner of overhead and buried sections, making the lightning wave propagation process more complex. The existing fixed-parameter grounding model is based on the assumption of uniform soil and cannot reflect the spatial differences and dynamic changes in the conductivity of mountainous soil. The fixed-weight current distribution method does not consider the local current pressure changes during the movement of thunderstorms. The passively triggered surge discharge method cannot match the steep wave propagation speed of mixed cable sections. The open-loop operation control mode does not feed back the actual on-site operating status to the model iteration stage. When applied to distributed wind power scenarios in mountainous areas, phenomena such as insufficient utilization of current dissipation capacity, overload of local grounding points, and lag in discharge action will occur, indicating room for further optimization and improvement. To solve this problem, we provide a grounding and lightning protection system and construction method for distributed wind power in mountainous areas. Summary of the Invention

[0003] The purpose of this invention is to provide a distributed wind power grounding and lightning protection system and construction method for mountainous areas, so as to solve the problems mentioned in the background art.

[0004] To achieve the above objectives, a mountain-based distributed wind power grounding and lightning protection system is provided, which is electrically connected to a mountain-based distributed wind turbine, a grounding hardware module with a controllable current dissipation branch, a cable lightning protection module with a controllable discharge path, and a field sensing module. The system includes a mountain grounding characteristic dynamic modeling unit, a grounding point current dissipation collaborative optimization unit, a cable transition lightning protection active control unit, and a multi-node state fusion management and control unit. The mountain grounding characteristic dynamic modeling unit receives terrain data, soil resistivity data, and lightning activity data, performs fusion calculations, generates a mountain current dissipation characteristic map, and outputs it synchronously to the grounding point current dissipation collaborative optimization unit and the cable transition lightning protection active control unit. The grounding point current dissipation collaborative optimization unit calculates the current dissipation weight allocation instructions for each grounding point based on the mountain current dissipation characteristic map and sends them to the grounding hardware module for pre-adjusting the conduction status of the controllable current dissipation branch. During the thunderstorm warning period, the pre-adjustment results of the current dissipation weight are simultaneously sent to the cable transition lightning protection active control unit. The active control unit for cable transition lightning protection combines the pre-adjustment results of current dissipation weight with real-time monitoring data to complete surge prediction calculation, outputs trigger commands to the cable lightning protection module in advance to conduct the corresponding discharge path, and sends the trigger execution status back to the dynamic modeling unit for mountain grounding characteristics. The multi-node status fusion control unit accesses the full node operation data to complete anomaly identification and early warning, obtains correction parameters, and sends the correction parameters back to the mountain grounding feature dynamic modeling unit to iteratively update the mountain runoff feature map.

[0005] The second objective of this invention is to provide a construction method for implementing a distributed wind power grounding and lightning protection system in mountainous areas, comprising the following steps: S1. Access topographic data, soil resistivity data, and lightning activity data. After preprocessing, perform fusion calculations to generate a mountain runoff characteristic map. S2. Calculate the current distribution weight allocation instruction for each grounding point based on the mountain current distribution characteristic map, send the current distribution weight allocation instruction to the grounding hardware module, complete the pre-adjustment of the conduction state of the controllable current distribution branch, determine the thunderstorm warning period through multi-source lightning data classification, and send the pre-adjustment result of the current distribution weight to the surge prediction stage during the thunderstorm warning period. S3. Combine the pre-adjustment results of the current dissipation weight with real-time monitoring data to complete the surge prediction calculation. Determine the advance sending time of the instruction based on the inherent conduction action duration of the controllable discharge path, output the trigger instruction to the cable lightning protection module in advance, sort the controllable discharge paths according to the current dissipation weight of the corresponding grounding point, complete the active guidance and discharge of lightning surge, and send the trigger execution status back to the multi-source data fusion modeling stage. S4. Access all node operation data to complete anomaly identification and early warning on the grounding side and lightning protection side. Extract corresponding correction parameters according to different anomaly types, and send the correction parameters back to the multi-source data fusion modeling stage to iteratively update the mountain runoff feature map.

[0006] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention addresses the problems of poor terrain adaptability of fixed parameter grounding models, uneven distribution of current dissipation capacity leading to single-point overload, passive and delayed surge discharge, and disconnect between the operating model and the actual field conditions in mountainous distributed wind power scenarios through dynamic current dissipation spectrum modeling with multi-factor correction, dynamic allocation of multi-constraint current dissipation weights, terrain-adaptive thunderstorm graded early warning, active prediction of frequency-related surges, and full-process active protection technology with multi-node state closed-loop iteration. It achieves dynamic optimization of grounding current dissipation capacity and precise active discharge of lightning surges under complex mountainous terrain, effectively reducing the probability of lightning strike failures in mountainous wind power and extending the service life of grounding and lightning protection equipment. Attached Figure Description

[0007] Figure 1 This is an overall block diagram of the present invention; Figure 2 This is the overall flowchart of the present invention. Detailed Implementation

[0008] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0009] Please see Figure 1 As shown, the present invention provides a mountain-based distributed wind power grounding and lightning protection system, which is electrically connected to a mountain-based distributed wind turbine, a grounding hardware module with a controllable current dissipation branch, a cable lightning protection module with a controllable discharge path, and a field sensing module. It includes a mountain grounding characteristic dynamic modeling unit, a grounding point current dissipation collaborative optimization unit, a cable transition lightning protection active control unit, and a multi-node state fusion management and control unit. The mountain grounding characteristic dynamic modeling unit receives terrain data, soil resistivity data, and lightning activity data, performs fusion calculations, generates a mountain current dissipation characteristic map, and outputs it synchronously to the grounding point current dissipation collaborative optimization unit and the cable transition lightning protection active control unit. The grounding point current dissipation collaborative optimization unit calculates the current dissipation weight allocation instructions for each grounding point based on the mountain current dissipation characteristic map and sends them to the grounding hardware module for pre-adjusting the conduction status of the controllable current dissipation branch. During the thunderstorm warning period, the pre-adjustment results of the current dissipation weight are simultaneously sent to the cable transition lightning protection active control unit. The active control unit for cable transition lightning protection combines the pre-adjustment results of current dissipation weight with real-time monitoring data to complete surge prediction calculation, outputs trigger commands to the cable lightning protection module in advance to conduct the corresponding discharge path, and sends the trigger execution status back to the dynamic modeling unit for mountain grounding characteristics. The multi-node status fusion control unit accesses the full node operation data to complete anomaly identification and early warning, obtains correction parameters, and sends the correction parameters back to the mountain grounding feature dynamic modeling unit to iteratively update the mountain runoff feature map.

[0010] The topographic data includes basic data on mountain elevation, slope, and aspect obtained from previous surveys, as well as actual topographic change data updated synchronously after each grounding construction, wind turbine location adjustment, and cable route change. Soil resistivity data is collected at fixed intervals by resistivity acquisition devices deployed at different soil depths at each grounding point. During the acquisition process, soil moisture content and soil temperature data at the corresponding points are acquired simultaneously to complete resistivity value correction. Lightning activity data consists of local electric field intensity and lightning proximity signals collected by atmospheric electric field acquisition devices and traveling wave monitoring devices deployed locally at the wind farm, and lightning strike location, lightning current amplitude, and lightning movement path data obtained from the regional lightning location system. All collected topographic data, soil resistivity data, and lightning activity data are preprocessed before being fused and calculated to generate a mountain diffuse current characteristic map.

[0011] Using the coverage area of ​​the 35MW distributed wind power project in Pinglu District as the modeling boundary, the basic data of mountain elevation, slope and aspect are imported into the modeling process. The actual terrain change data updated synchronously after each grounding construction, wind turbine location adjustment and cable route change are superimposed to generate a three-dimensional mountain terrain grid model covering the entire wind field. During the modeling process, the wind field area is divided into square grid units with a side length of ten meters. Each grid unit corresponds to a unique spatial coordinate and elevation value. Such grid division accuracy can match the deployment density of grounding electrodes and sensors, avoiding feature mapping errors caused by excessively large grid scale.

[0012] After completing the construction of the 3D mountain terrain mesh model, the generation of the soil resistivity distribution layer for the entire region began. The numerically corrected soil resistivity data, combined with the soil moisture content and soil temperature at corresponding points, were mapped to the corresponding grid positions in the 3D mountain terrain mesh model according to the coordinates of the collected points. During the correction process, mountain slope correction terms and soil moisture infiltration lag correction terms, distinct from those in existing flat terrain models, were added. The corrected soil resistivity calculation formula is as follows: , in the formula coordinates The grid is in The corrected soil resistivity at time t, in ohm-meters. The original soil resistivity, measured in ohm-meters, is collected by sensors at different burial depths at corresponding locations. This is a soil temperature correction coefficient, a dimensionless value, whose value varies with soil temperature. The correction factor is set to 1.25 at 0 degrees Celsius, 1 at 25 degrees Celsius, and decreases linearly with increasing temperature above 25 degrees Celsius. Here is the lag correction coefficient for soil moisture content, which is a dimensionless value. This represents the real-time soil moisture content at the corresponding location, expressed as a percentage. The lag time is the time it takes for rainfall to infiltrate to the depth of the sensor, measured in hours. This lag time is calculated based on the grid slope and soil texture; a steeper slope results in a longer lag time. This is the slope gradient correction factor for runoff in mountainous terrain; it is a dimensionless value. The ground slope corresponding to the grid is measured in degrees. When the slope is zero degrees, the correction factor is 1. As the slope increases, the correction factor increases linearly with the extension of the diffusion path. The soil resistivity obtained by this correction can truly reflect the resistivity changes caused by the undulation of the mountainous terrain and the lag of water infiltration, avoiding the calculation deviation of diffusion capacity caused by direct mapping of flat land models.

[0013] After completing the resistivity correction of all grids, a full-area soil resistivity distribution layer can be generated. Each grid in the layer corresponds to a unique corrected resistivity value, which will be directly used for subsequent calculation of the current dissipation capacity weight.

[0014] After generating the soil resistivity distribution layer for the entire region, the generation of the regional lightning risk probability distribution layer begins. This involves combining local electric field strength and nearby lightning signals with lightning strike location, lightning current amplitude, and lightning path data. The probability of lightning occurrence at each grid point is calculated by matching spatial coordinates with time series data. A skin effect correction term, distinct from existing technologies, is incorporated into the calculation process to avoid the problem that conventional power frequency resistance models cannot adequately reflect the reduced current dissipation capacity under high-frequency lightning impacts. The formula for calculating the lightning occurrence probability is as follows: , in the formula coordinates The grid is in The probability of lightning occurring at any given time ranges from 0 to 1 and is a dimensionless value. The lightning density for the corresponding location is calculated based on historical lightning strike locations and lightning movement paths, with the dimension being lightning strikes per square kilometer per year. The equivalent interception area of ​​a single grid cell is expressed in square kilometers. This is the skin effect correction coefficient for lightning current, which is a dimensionless value. To monitor the lightning current amplitude in real time, the dimension is kiloampere. The larger the lightning current amplitude, the higher the frequency component. The smaller the soil skin depth and the smaller the equivalent current dissipation area, the larger the correction coefficient value. During the calculation process, the local electric field strength and the lightning proximity signal are simultaneously connected to correct the probability value in real time. When the electric field strength exceeds the threshold, the probability value of the corresponding area is increased proportionally.

[0015] After calculating the lightning occurrence probability for all grids, a regional lightning risk probability distribution layer can be generated. Each grid in the layer corresponds to a unique lightning occurrence probability value, which will be directly used for subsequent lightning protection level classification.

[0016] After generating the two distribution layers, the soil resistivity distribution layer of the whole region and the lightning risk probability distribution layer of the region are spatially aligned grid by grid. During the alignment process, the grid coordinates and scales of the two layers are completely consistent and there is no coordinate offset. After the alignment is completed, each grid point is assigned a corresponding current dissipation capacity weight and lightning protection level. The current dissipation capacity weight is calculated based on the soil resistivity after grid correction. The lower the resistivity, the greater the current dissipation capacity weight. The current dissipation capacity weight ranges from 0 to 1 and is a dimensionless value. The lightning protection level is divided according to the probability of lightning occurrence in the grid. The higher the probability, the higher the protection level. The protection level is increased step by step from 1 to 4 by integer values.

[0017] Once the assignment is complete, a dynamic mountain runoff characteristic map will be generated, which will change with the updates of the collected data. After the map is generated, it will be output synchronously to the subsequent runoff weight calculation and surge prediction stages. The runoff capacity weight will be used as the core basis for the runoff weight allocation of the grounding point, and the lightning protection level will be used as the core basis for the setting of the thunderstorm warning threshold and the calculation of the advance conduction time of the discharge path. Each time new sensor data is received, the above modeling process will be repeated to update the map, ensuring that the map can truly reflect the dynamic changes of mountain soil and lightning characteristics.

[0018] Conventional flat-ground grounding modeling methods directly map the measured soil resistivity onto a planar grid, without considering the extended runoff path caused by the mountain slope and the lightning attraction effect at the protruding ridge. They also fail to consider the differences in runoff and infiltration time of water on different slopes after rainfall, and do not consider the skin effect of high-frequency components of lightning current in the soil. This results in a large deviation between the generated runoff characteristic map and the actual situation in the mountains, making it difficult to support subsequent dynamic allocation of runoff weights and active lightning protection control. Based on this, a specific construction process for the mountain runoff characteristic map is carried out.

[0019] Using the 12.6 square kilometer area covered by the 35MW distributed wind power project in Pinglu District as the modeling boundary, the basic data of mountain elevation, slope, and aspect obtained from previous surveys were imported into the modeling process. The WGS84 coordinate system was used as a unified spatial reference during data import to avoid mapping deviations caused by inconsistent coordinates from multiple data sources. Actual terrain change data, updated synchronously after each grounding construction, wind turbine location adjustment, and cable route change, was overlaid. This actual terrain change data includes the terrain backfill height after the grounding electrode was buried, the terrain elevation changes after the wind turbine foundation was poured, and the terrain slope changes after the cable trench excavation and backfilling, generating a mountainous area covering the entire wind farm. The three-dimensional terrain mesh model divides the wind field area into square grid cells with a side length of ten meters during the modeling process. Each grid cell corresponds to a unique latitude and longitude coordinate, elevation value, slope value, and aspect value. The ten-meter grid division accuracy can match the average fifty-meter spacing between the grounding point and the sensor. Each grid contains at least one sensor acquisition point or grounding point to avoid feature smoothing errors caused by excessively large grid scales and to avoid computational redundancy caused by excessively small grid scales. The generated three-dimensional mountain terrain mesh model will serve as the spatial basis for all subsequent data mappings. All multi-source data mappings are performed using the grid center coordinates as the reference for spatial matching.

[0020] After completing the construction of the 3D mountain terrain mesh model, the generation of the soil resistivity distribution layer for the entire region was initiated. The numerically corrected soil resistivity data, combined with the soil moisture content and soil temperature at the corresponding points, were mapped to the corresponding grid positions in the 3D mountain terrain mesh model according to the coordinates of the collection points. The soil resistivity data came from resistivity acquisition devices deployed at three burial depths of 0.5 meters, 1 meter, and 2 meters at each grounding point, collected at ten-minute intervals. The soil moisture content and soil temperature came from sensors at the same locations and burial depths as the resistivity acquisition devices. During the mapping process, for grids without direct acquisition points, the inverse distance weighted interpolation method was used to calculate the resistivity value of the corresponding grid by combining the values ​​of adjacent acquisition points. In the interpolation process, the weight was inversely proportional to the square of the distance from the grid to the acquisition point to avoid interference from the values ​​of distant acquisition points on the interpolation results.

[0021] Four correction terms, distinct from existing flatland models, were added during the calibration process: soil temperature correction, soil moisture lag correction, mountain slope runoff correction, and slope surface runoff correction. The corrected formula for calculating the resistivity of stratified soil is as follows: , in the formula coordinates Grid burial depth Location at The corrected soil resistivity at time t, in ohm-meters, represents the corresponding burial depth. The original soil resistivity, measured in ohmmeters, is collected by the sensor at the location. This is the soil temperature correction coefficient, a dimensionless value, whose value varies with burial depth. Soil temperature at the location The correction factor is set to 1.25 at 0 degrees Celsius and 1 at 25 degrees Celsius. The correction factor increases by 0.02 for every degree Celsius below 0 degrees Celsius and decreases by 0.01 for every degree Celsius above 25 degrees Celsius. This correction factor accurately reflects the characteristic of increased soil resistivity under low-temperature freezing conditions. Here is the lag correction coefficient for soil moisture content, which is a dimensionless value. Burial depth at corresponding points Place Real-time soil moisture content at any given time, expressed as a percentage. To allow water to seep into the burial depth after rainfall Location lag time, in hours, varies with grid slope. As the slope increases, the lag time at a depth of 0.5 meters is taken as 2 hours when the slope is 0 degrees and as 6 hours when the slope is 30 degrees. This means that as the slope increases, rainwater runoff along the slope increases, the amount of infiltration decreases, and the time it takes for water to infiltrate to the same depth is longer. This is the slope gradient correction factor for runoff in mountainous terrain; it is a dimensionless value. The ground slope corresponds to the grid, measured in degrees. A correction factor of 1 is used when the slope is zero degrees, and increases by 0.08 for every five degrees increase in slope. As the slope increases, the actual path length of the current dissipating along the slope is greater than the projected length on the plane, thus increasing the current dissipation resistance. This is the slope runoff correction factor, which is a dimensionless value. The angle between the grid slope aspect and the rainfall wind direction is expressed in degrees. When the slope aspect is in the same direction as the rainfall wind direction, the amount of rainfall received by the slope increases and the moisture content rises. The correction factor is 0.9. When the slope aspect is opposite to the rainfall wind direction, the amount of rainfall received by the slope decreases. The correction factor is 1.1.

[0022] Taking a specific grid as an example, the corresponding grid is located on an east-facing slope with a gradient of 20 degrees. During rainfall, the wind direction is east. The original resistivity measurement at a depth of 0.5 meters is 120 ohm-meters. The soil temperature is 10 degrees Celsius. The time remaining until the end of rainfall is 4 hours. Substituting these values ​​into the formula, the temperature correction factor is 1.06, the hysteresis correction factor is 1.03, the slope correction factor is 1.32, and the runoff correction factor is 0.9. The final corrected resistivity is 120 × 1.06 × 1.03 × 1.32 × 0.9 ≈ 156 ohm-meters. The corrected resistivity is 30% higher than the original measurement value, which can accurately reflect the effect of the extended runoff path caused by the slope and avoid the problem of underestimating the runoff resistance of mountainous areas in flatland models.

[0023] After completing the resistivity correction for all grids at three burial depths, the resistivity at different burial depths is weighted and averaged with weights of 0.5, 0.3, and 0.2 to obtain the equivalent resistivity of the grid, generating a soil resistivity distribution layer for the entire region. Each grid in the layer corresponds to a unique equivalent resistivity value, which will be directly used for subsequent current dissipation capacity weight calculation. The layered resistivity values ​​at different burial depths will be used for subsequent selection of grounding electrode burial depth and formulation of current dissipation branch conduction strategies.

[0024] After generating the soil resistivity distribution layer for the entire region, the generation of the regional lightning risk probability distribution layer begins. The local electric field strength and lightning proximity signals are combined with lightning strike location, lightning current amplitude, and lightning movement path data. The probability of lightning occurrence at each grid point is calculated by matching spatial coordinates with time series. The local electric field strength is obtained from the values ​​collected by the atmospheric electric field acquisition device deployed at the commanding height of the wind farm at a one-minute cycle. The lightning proximity signal is obtained from the lightning leader signal collected by the traveling wave monitoring device. The lightning strike location, lightning current amplitude, and lightning movement path data are obtained from the connected regional lightning location system. The data update cycle is one minute.

[0025] The calculation process incorporates three correction terms that differ from existing technologies: a skin effect correction term for lightning current, a terrain-induced lightning correction term, and a ground-object shielding correction term. The formula for calculating the probability of lightning occurrence is as follows: , in the formula coordinates The grid is in The probability of lightning occurring at any given time ranges from 0 to 1 and is a dimensionless value. The rolling lightning density for the corresponding location is calculated based on historical lightning strike locations and real-time lightning movement path data from the past ten years. The dimension is times per square kilometer per year. During real-time calculation, the predicted lightning density values ​​for different grids are dynamically updated according to the lightning movement path and speed. The equivalent interception area for a single grid cell is expressed in square kilometers. The equivalent interception area increases with increasing grid elevation. This is the skin effect correction coefficient for lightning current, which is a dimensionless value. The lightning current amplitude is measured in kiloamperes (kA) for real-time monitoring. A larger lightning current amplitude indicates a higher proportion of high-frequency components, a smaller skin depth, and a smaller equivalent current dissipation area. For a lightning current amplitude of 10 kA, the skin depth is approximately 5 meters, with a correction factor of 1.1. For a lightning current amplitude of 100 kA, the skin depth is approximately 0.5 meters, with a correction factor of 1.8. Here is the terrain-based mine-attracting correction factor, which is a dimensionless value. The grid's protrusion height relative to the average terrain within a 100-meter radius is expressed in meters. A correction factor of 1 is applied when the protrusion height is 0, and increases by 0.15 for every 10-meter increase in protrusion height. Prominent locations such as ridges are more likely to generate upward leaders and have a higher probability of being struck by lightning. is the ground cover shielding correction factor, which is a dimensionless value. The correction factor is 0.3 when the grid is within the protection range of the wind turbine blade lightning arrester, 0.6 when it is within the shielding range of the wind turbine tower, and 1 when there is no shielding.

[0026] Taking a specific ridge grid as an example, the corresponding grid is located at a prominent point on the ridge, with a protrusion height of 30 meters. It is outside the protection range of the wind turbine lightning arrester, with a rolling lightning density of 4 times per square kilometer per year. The equivalent interception area is 0.0001 square kilometers. The real-time monitored lightning current amplitude is 50 kA. The skin effect correction factor is 1.5, the terrain-induced lightning correction factor is 1.45, and the shielding correction factor is 1. Substituting these values ​​into the formula, the probability of lightning occurrence is calculated as follows: This translates to a lightning strike probability of 0.087%, which is 45% higher than the calculation result without considering terrain and skin correction, accurately reflecting the high lightning strike risk at the ridge location.

[0027] During the calculation process, the probability value is corrected in real time by synchronously accessing the local electric field strength and the lightning proximity signal. When the electric field strength exceeds 10 kV per meter, the probability value of the corresponding area is increased to twice the original value. When the lightning leader development signal is detected, the probability value of the grid on the corresponding path is increased to five times the original value.

[0028] After calculating the lightning occurrence probability for all grids, a regional lightning risk probability distribution layer is generated. Each grid in the layer corresponds to a unique lightning occurrence probability value. The lightning occurrence probability value will be directly used for subsequent lightning protection level classification. The skin correction coefficient corresponding to different amplitude lightning currents will be used for subsequent discharge path number calculation.

[0029] After generating the two distribution layers, the soil resistivity distribution layer for the entire region and the lightning risk probability distribution layer for the region are spatially aligned grid by grid. During the alignment process, the coordinates of the mountain 3D terrain mesh model are used as the reference, and sub-pixel level registration is performed on the mesh boundaries of the two layers to ensure that the mesh coordinates and scales of the two layers are completely consistent, without any coordinate offset or scale deviation. After alignment, each mesh point is assigned a corresponding current dissipation capacity weight and lightning protection level. The current dissipation capacity weight is calculated based on the equivalent resistivity of the mesh; the lower the equivalent resistivity, the greater the current dissipation capacity weight. For equivalent resistivity less than 100... The current dissipation capacity weight is 1 when the equivalent resistivity is greater than 1000 ohm-meters, and 0 when the equivalent resistivity is greater than 1000 ohm-meters. Intermediate values ​​are calculated by linear interpolation. The current dissipation capacity weight ranges from 0 to 1 and is a dimensionless value. The lightning protection level is divided according to the probability of lightning occurrence in the grid. When the probability is greater than 0.001, the protection level is level 4; when the probability is between 0.0005 and 0.001, the protection level is level 3; when the probability is between 0.0001 and 0.0005, the protection level is level 2; and when the probability is less than 0.0001, the protection level is level 1. The protection level increases progressively from 1 to 4 using integer values.

[0030] After the assignment is complete, a dynamic mountain runoff characteristic map will be generated, which will update with the collected data. Once generated, the map will be synchronously output to the subsequent runoff weight calculation and surge prediction stages. The runoff capacity weight will directly serve as the core basis for the runoff weight allocation at grounding points. During runoff weight allocation, more runoff tasks will be assigned to grounding points near grids with high runoff capacity weights, while no runoff tasks will be assigned near grids with a runoff capacity weight of 0. Layered resistivity values ​​will be used to guide the runoff branch conduction strategy for grounding electrodes at different burial depths. When the surface resistivity is high, runoff branches of deeply buried grounding electrodes will be prioritized. The lightning protection level will be directly used as the basis for... This serves as the core basis for setting thunderstorm warning thresholds and calculating the advance conduction time of discharge pathways. In areas with a protection level of 4, the thunderstorm warning threshold is reduced by 30%, and the advance conduction time of discharge pathways is increased by 50%, ensuring that surges in high-risk areas can be discharged quickly. After each new sensor data is received, the above modeling process is repeated to update the map. Under normal weather conditions, the update cycle is consistent with the sensor acquisition cycle at ten minutes. During thunderstorm warning periods, the update cycle is shortened to one minute, ensuring that the map can truly reflect the dynamic changes in mountain soil and lightning characteristics, providing accurate spatial data support for subsequent dynamic adjustment of runoff and active lightning protection.

[0031] Conventional grounding current distribution uses a fixed weighting method, which does not consider the dynamic changes in soil resistivity in mountainous areas and the dynamic effects of thunderstorm movement. This easily leads to overload of grounding points in high-risk areas and idle grounding points in low-risk areas, making it difficult to fully utilize the current dissipation capacity of the distributed grounding network. Based on this, a dynamic calculation and instruction generation process for current dissipation weights is developed. When calculating current dissipation weights, parameters are first extracted from the mountainous current dissipation characteristic map. The extraction process takes the actual latitude and longitude coordinates of each grounding point as the center and takes the current dissipation capacity weights of the surrounding 3×3 grids (9 grids) and the lightning protection level as a weighted average. This avoids calculation deviations caused by errors in individual grid data. In the weighting process, the weight value of each grid is inversely proportional to the square of the distance from the grounding point to the grid center; the closer the distance, the higher the weight value. Finally, the average current dissipation capacity weight and average lightning protection level corresponding to each grounding point are obtained. The average current dissipation capacity weight is used for subsequent calculation of the maximum dissipable current dissipation ratio, and the average lightning protection level is used for subsequent determination of current dissipation priority.

[0032] After parameter extraction, the maximum distributable current dissipation ratio for each grounding point is calculated. Three correction terms, distinct from existing fixed-capacity calculations, are incorporated into the calculation: grounding electrode temperature correction, grounding electrode corrosion margin correction, and soil current dissipation capacity correction. The formula for calculating the maximum distributable current dissipation ratio is as follows: , in the formula For the first One grounding point at The maximum distributable current distribution ratio at any given time, ranging from 0 to 1, is a dimensionless value. For the first The total number of controllable current-dissipating branches connected to each grounding point, in units of branches. This refers to the inherent rated current capacity of a single controllable current-dissipating branch, expressed in kiloamperes (kA). The inherent rated current capacity is the long-term permissible current value specified at the factory when the controllable current-dissipating branch leaves the factory. This is the grounding electrode temperature correction factor, which is a dimensionless value. For the first The real-time temperature of the grounding electrode at each grounding point, in degrees Celsius. A correction factor of 1 is applied when the temperature is 25 degrees Celsius. The correction factor decreases by 0.05 for every 10 degrees Celsius increase in temperature. Increased temperature leads to increased grounding electrode resistance and decreased current-carrying capacity. This is the corrosion allowance coefficient for the grounding electrode, which is a dimensionless value. For the first The service life of each grounding electrode at each grounding point is specified in years. A correction factor of 1 is applied to newly commissioned grounding electrodes, decreasing by 0.02 for each additional year of service. Long-term operation of grounding electrodes can lead to surface corrosion, resulting in a decrease in the effective current-carrying cross-section. This is a correction factor for soil runoff capacity, and it is a dimensionless value. For the first The average corrected soil resistivity around each grounding point is expressed in ohm-meters. The correction factor is set to 1 when the average current dissipation capacity weight is 1. For every 0.1 decrease in the current dissipation capacity weight, the correction factor decreases by 0.1. An increase in soil resistivity leads to an increase in current dissipation resistance and a decrease in current carrying capacity. for The expected total lightning current amplitude at any given time, in kiloamperes, is determined based on the real-time monitored lightning current amplitude and historical lightning current statistics.

[0033] Taking the actual calculation of a wind turbine grounding point as an example, the corresponding grounding point is connected to 10 controllable current dissipation branches. The inherent rated current capacity of a single branch is 2 kA, the real-time temperature of the grounding electrode is 40 degrees Celsius, the temperature correction factor is 0.95, the service life of the grounding electrode is 5 years, the corrosion margin factor is 0.9, the weight of the surrounding average current dissipation capacity is 0.8, the soil current dissipation capacity correction factor is 0.8, and the real-time expected total lightning current amplitude is 100 kA. Substituting into the formula, the maximum dissipable current dissipation ratio is calculated to be (10×2×0.95×0.9×0.8) / 100=0.1368, which means that the corresponding grounding point can bear a maximum of 13.68% of the total lightning current dissipation task. The calculated maximum dissipable current dissipation ratio will be used as the upper limit of the subsequent current dissipation weight allocation. Under any circumstances, the current dissipation weight allocated to a single grounding point cannot exceed the corresponding value to avoid overload of the grounding point and potential rise.

[0034] After calculating the maximum distributable current dissipation ratio, the current dissipation priority of each grounding point is determined based on the lightning protection level of the corresponding grid location. A distance correction term, distinct from existing methods that only rank by risk level, is incorporated into the calculation. The closer the electrical distance from the grounding point to the junction of the overhead and buried cables, the shorter the current dissipation path, the faster the surge discharge, and the higher the priority. The formula for calculating the current dissipation priority is as follows: , in the formula For the first One grounding point at The current propagation priority at any given time; higher values ​​indicate higher priority, and are dimensionless. For the first The average lightning protection level of the grid surrounding each grounding point, taking values ​​from 1 to 4 (integers). This is the distance correction factor, which is a dimensionless value. For the first The electrical distance from the grounding point to the cable connection is given in meters. When the electrical distance is 0, the correction factor is 1. The correction factor decreases by 0.1 for every 100 meters increase in distance. When the electrical distance exceeds 1000 meters, the correction factor is 0.

[0035] Taking two grounding points as an example, the average lightning protection level around the first grounding point is level 4, and it is 50 meters away from the cable connection. The distance correction factor is 0.95, and the current dissipation priority is 4×0.95=3.8. The average lightning protection level around the second grounding point is level 3, and it is 200 meters away from the cable connection. The distance correction factor is 0.8, and the current dissipation priority is 3×0.8=2.4. The first grounding point has a higher priority than the second grounding point. When allocating current dissipation weights, the current dissipation needs of the higher priority grounding point should be given priority.

[0036] After the current dissipation priority is sorted, under normal circumstances, the current dissipation weight is evenly distributed according to the maximum dissipable current dissipation ratio of each grounding point. The distribution process proceeds in descending order of priority. First, the grounding point with the highest priority is assigned a weight value not exceeding its own maximum dissipable current dissipation ratio. After the high-priority grounding points are assigned, the weights are then assigned to the grounding points with the next lower priority, until the total weight of all grounding points is 1, meaning that all lightning currents are borne by a corresponding grounding point. If the maximum dissipable ratio of a high-priority grounding point has been used up during the distribution process, the remaining current dissipation weight is automatically allocated to the grounding point with the next lower priority, ensuring that the weight of all grounding points does not exceed their own maximum dissipable current dissipation ratio and that no overload occurs.

[0037] During thunderstorm warning periods, the initially allocated current dissipation weights are dynamically adjusted based on lightning movement path data. The adjustment process first identifies grounding points within the lightning movement path's coverage area and calculates the difference between the current allocated current dissipation weight and the maximum dissipable current dissipation ratio for each grounding point. A negative difference indicates that the allocated weight exceeds the grounding point's carrying capacity, posing an overload risk. The excess current dissipation weight needs to be transferred to surrounding grounding points. During the weight transfer process, an electrical distance correction term and a remaining capacity matching term, distinct from the existing nearest-neighbor transfer, are added. Priority is given to transferring to grounding points with close electrical distance and sufficient remaining capacity. The formula for calculating the weight transfer amount is as follows: , in the formula overload grounding point It needs to be transferred to the surrounding grounding point. The divergence weight is a dimensionless value. Grounding point Excess flow weights exceeding the maximum allocatable proportion are dimensionless values. is the remaining allocable capacity of the grounding point, is the difference between the maximum allocable current dissipation ratio and the current allocation weight, and is a dimensionless value. Grounding point With grounding point The electrical distance correction factor is a dimensionless value. The closer the electrical distance between the two grounding points, the higher the correction factor, and the smaller the loss during current dissipation. This refers to the set of grounding points that are not within the lightning strike range and have remaining capacity.

[0038] Taking an overloaded grounding point as an example, the current weight assigned to the grounding point is 0.18, the maximum allocable current dissipation ratio is 0.13, and the excess current dissipation weight is 0.05. There are three grounding points in the vicinity that are not covered by lightning and have remaining capacity, with remaining capacities of 0.06, 0.04, and 0.03 respectively, and electrical distance correction coefficients of 0.9, 0.8, and 0.7 respectively. Substituting into the formula, the weights received by the three grounding points are calculated to be 0.0225, 0.02, and 0.0075 respectively. After the transfer is completed, the weight assigned to all grounding points does not exceed their own maximum allocable current dissipation ratio, and there will be no single-point current dissipation overload. After the weight transfer is completed, the final current dissipation weights of each grounding point during the thunderstorm period are obtained.

[0039] After obtaining the final determined current dissipation weights for each grounding point, the corresponding controllable current dissipation branch conduction quantity and conduction duration control parameters are generated. During the calculation process, a redundancy correction term, different from the existing proportional conduction, is added to reserve a certain proportion of spare branches to cope with branch faults or unexpected lightning current amplitudes. The formula for calculating the conduction quantity is as follows: , in the formula For the first The number of controllable current-dissipating branches that need to be connected for each grounding point, expressed in terms of the number of branches, with the result rounded up to the nearest integer. For the first The final current dissipation weight assigned to each grounding point is a dimensionless value. The redundancy factor is set to 1.2, with 20% redundancy capacity reserved. The conduction time is determined based on the expected duration of the thunderstorm warning period, from the start of the thunderstorm warning to 10 minutes after the thunderstorm warning ends, ensuring that some diffuser branches are closed only after the thunderstorm has completely passed.

[0040] Taking the previously calculated grounding point as an example, the final current distribution weight of the corresponding grounding point is 0.1368, the expected lightning current amplitude is 100 kA, the rated current capacity of a single branch is 2 kA, and the redundancy coefficient is 1.2. Substituting into the formula, the number of branches that need to be connected is ceil((0.1368×100 / 2)×1.2)=9, with the remaining branch as a backup. The connection time lasts from the start of the thunderstorm warning to 10 minutes after the end of the thunderstorm.

[0041] After determining the number of grounding points, their conduction duration, and the number of the conducting branches, all parameters are combined to form a current dissipation weight allocation instruction. This instruction is then sent to the grounding hardware module via a communication link. Upon receiving the instruction, the grounding hardware module adjusts the conduction status of the controllable current dissipation branches according to the parameters. During thunderstorm warning periods, while issuing the current dissipation weight allocation instruction, a pre-adjustment result for the current dissipation weight is simultaneously generated. This result includes the final allocation weight for each grounding point, the number of pre-conducting branches, and the total discharge capacity. The pre-adjustment result for the current dissipation weight is then sent to the cable transition lightning protection active control unit for use in subsequent surge prediction calculations. This ensures that the surge prediction process can accurately grasp the actual total discharge capacity of the grounding network, avoiding overvoltage caused by insufficient discharge capacity.

[0042] This dynamic weight allocation method can improve the current dissipation capacity utilization of the distributed grounding grid by more than 30%, avoid ground potential backflash caused by single-point overload, and send pre-adjustment results to the lightning protection unit in advance, so that the coordination between the grounding grid and the cable lightning protection device can be more precise during surge discharge, effectively reduce the overvoltage level at the cable connection, and improve the lightning protection reliability of mountain distributed wind power.

[0043] Conventional grounding current distribution often uses a fixed weight ratio, which does not take into account the differences in current distribution paths caused by the undulation of mountainous terrain and the dynamic changes in soil conditions, nor does it make dynamic adjustments in conjunction with the movement of thunderstorms. This can easily lead to problems such as overload of grounding points in high-risk areas and idle current distribution capacity of grounding points in low-risk areas. It is difficult to fully utilize the overall current distribution capacity of the distributed grounding network, and there is also the risk of grounding potential rising and backflashover equipment. Based on this, we carried out a full-process implementation of dynamic calculation of current distribution weight and command generation.

[0044] When calculating the current diffusion weight, basic parameters are first extracted from the mountain current diffusion characteristic map. The extraction process uses the actual latitude and longitude coordinates of each grounding point as the center, and takes a distance-weighted average of the current diffusion capacity weights and lightning protection levels of all grids within a 20-meter radius. This avoids calculation bias caused by anomalies in individual grid data. During the weighting process, the weight value of each grid is inversely proportional to the straight-line distance from the grid center to the grounding point; the closer the distance, the higher the weight value. Finally, the regional average current diffusion capacity weight and regional average lightning protection level are obtained for each grounding point. The regional average current diffusion capacity weight is used for subsequent calculation of the maximum distributable current diffusion ratio, and the regional average lightning protection level is used for subsequent current diffusion priority determination. After completing the basic parameter extraction, the maximum distributable current diffusion ratio for each grounding point is calculated. During the calculation, three dynamic correction terms, different from the existing direct conversion method of rated capacity, are added: grounding electrode temperature correction term, soil condition correction term, and slope path correction term. The formula for calculating the maximum distributable current diffusion ratio is as follows: , in the formula For the first One grounding point at The maximum distributable current distribution ratio at any given time, ranging from 0 to 1, is a dimensionless value. For the first The total number of controllable current-dissipating branches connected to each grounding point, in units of branches. This refers to the inherent rated current capacity of a single controllable current-dissipating branch, expressed in kiloamperes (kA). The inherent rated current capacity is the long-term permissible current capacity determined by type testing at the time of manufacture of the controllable current-dissipating branch. This is the grounding electrode temperature correction factor, which is a dimensionless value. For the first The real-time operating temperature of the grounding electrode at each grounding point is measured in degrees Celsius. A correction factor of 1 is applied at 25 degrees Celsius, decreasing by 0.06 for every 10 degrees Celsius increase in temperature. When a large current flows through the grounding electrode, causing its temperature to rise, the grounding electrode resistance will increase accordingly, and its current-carrying capacity will decrease. This is a soil condition correction factor, which is a dimensionless value. To reflect the real-time state of the soil around the grounding point, a correction factor of 1 is used when the soil is moist and not frozen, 0.8 when the soil is dry, and 0.7 when the surface of the soil is frozen. Soil dryness or freezing will increase resistivity and reduce current dissipation efficiency. This is the slope path correction factor, which is a dimensionless value. This represents the ground slope at the location of the grounding point, in degrees. The correction factor is 1 when the slope is 0 degrees, and decreases by 0.05 for every 5 degrees increase in slope. As the slope increases, the actual path length of the lightning current along the slope is greater than the horizontal projected length, and the current dissipation resistance will increase accordingly. for The maximum lightning current amplitude predicted at all times based on lightning monitoring data, in kiloamperes.

[0045] Taking the actual calculation of a grounding point at a ridge location in the Pinglu project as an example, the corresponding grounding point is connected to a total of 8 controllable current dissipation branches. The inherent rated current capacity of a single branch is 2 kA. The real-time monitoring grounding electrode temperature is 35 degrees Celsius, with a temperature correction factor of 0.94. The soil has been dry for 10 consecutive days without rainfall, with a soil condition correction factor of 0.8. The ground slope at the location is 25 degrees, with a slope path correction factor of 0.75. The predicted maximum lightning current amplitude is 100 kA. Substituting into the formula, the maximum dissipable current dissipation ratio is calculated to be (8×2×0.94×0.8×0.75) / 100=0.09024. That is, the corresponding grounding point can bear a maximum of 9.024% of the total lightning current discharge task. The calculated maximum dissipable current dissipation ratio will be used as the hard upper limit of the current dissipation weight allocation. Under any working condition, the current dissipation weight allocated to a single grounding point must not exceed the corresponding value to avoid overloading the grounding point and causing the ground potential to rise, which could then trigger the blower or cable equipment.

[0046] After calculating the maximum distributable current dissipation ratio, the current dissipation priority of each grounding point is determined. During the calculation, two distance correction terms are added, distinct from the existing method of sorting only by risk level: a tower base distance correction term and a cable junction point distance correction term. Grounding points closer to the tower base or cable junction point have shorter current dissipation paths, faster current discharge rates, and higher priority. The formula for calculating current dissipation priority is as follows: , in the formula For the first One grounding point at The current propagation priority at any given time; a higher value indicates a higher priority. This value is dimensionless. For the first The average lightning protection level of the area corresponding to each grounding point is an integer ranging from 1 to 4. This is the correction factor for the distance from the base of the tower, and it is a dimensionless value. This represents the electrical distance from the grounding point to the corresponding wind turbine tower base, in meters. A correction factor of 1.2 is used when the distance is 0 meters, and decreases by 0.05 for every 20 meters increase in distance. This is a correction factor for the distance at cable junction points; it is a dimensionless value. The distance from the grounding point to the connection point of the corresponding overhead cable and buried cable is expressed in meters. When the distance is 0, the correction factor is 1.1. For every 30 meters increase in distance, the correction factor decreases by 0.03.

[0047] Taking the actual calculation of two grounding points within the project as an example, the first grounding point is located next to the wind turbine tower base, with an average lightning protection level of 4 in the area. It is 5 meters away from the tower base, with a tower base distance correction factor of 1.1875. It is 20 meters away from the cable connection point, with a cable connection point distance correction factor of 1.08. The calculated current dissipation priority is 4 × 1.1875 × 1.08 ≈ 5.13. The second grounding point is located below the hillside, with an average lightning protection level of 2 in the area. It is 80 meters away from the tower base, with a tower base distance correction factor of 1.0. It is 150 meters away from the cable connection point, with a cable connection point distance correction factor of 0.85. The calculated current dissipation priority is 2 × 1.0 × 0.85 = 1.7. When allocating current dissipation weights, the current dissipation needs of the higher priority grounding point will be given priority.

[0048] After prioritizing the current dissipation, the current dissipation weight allocation under normal operating conditions is carried out. The allocation process proceeds sequentially from highest to lowest priority. First, the highest priority grounding point is allocated a weight value not exceeding its maximum dissipable current dissipation ratio. After the high-priority grounding points are allocated, the next lower priority grounding points are then allocated weights. During the allocation process, the allocation weight of the high-priority grounding point is 90% of its maximum dissipable current dissipation ratio, with a 10% redundancy capacity reserved to cope with sudden faults, until the total allocation weight of all grounding points is 1, meaning that all fault currents or lightning currents have corresponding discharge channels. After the allocation is completed, each grounding point's allocation weight is checked point by point to see if it exceeds the maximum dissipable current dissipation ratio. If it does, the excess weight is transferred to the next priority grounding point with remaining capacity to ensure that all grounding points operate within the safe capacity range. The current dissipation weights allocated under normal operating conditions will serve as the control basis for the number of controllable current dissipation branches in daily operation. The corresponding number of branches are kept in operation daily to meet the discharge requirements of power frequency grounding faults and small-amplitude lightning currents.

[0049] During thunderstorm warning periods, the initially allocated current dissipation weights are dynamically adjusted based on lightning path data. The adjustment process first identifies grounding points within the lightning path's coverage area. The arrival time of the thunderstorm at the corresponding grounding point is calculated based on the lightning's speed. A gradient transfer coefficient is set according to the arrival time: 0.3 for arrival times greater than 10 minutes, 0.6 for arrival times between 5 and 10 minutes, and 1 for arrival times less than 5 minutes. The difference between the current allocated weight and the maximum dissipable current dissipation ratio for each grounding point is calculated. A negative difference indicates an overload risk at the grounding point during a direct thunderstorm strike, requiring the transfer of excess current dissipation weight to surrounding grounding points. A mutual resistance correction term, distinct from the existing nearest-neighbor transfer, is added during the weight transfer process. The closer the electrical distance between two grounding points, the smaller the mutual resistance, resulting in a stronger current shielding effect during current dissipation and a lower transferable weight ratio. The formula for calculating the weight transfer amount is as follows: , in the formula overload grounding point It needs to be transferred to the surrounding grounding point. The divergence weight is a dimensionless value. Grounding point Excess current distribution weights exceeding the maximum allocatable current distribution ratio are dimensionless values. Here, represents the time gradient transfer coefficient, which is a dimensionless value. Grounding point The remaining allocable capacity is the difference between the maximum allocable current distribution ratio and the current allocation weight, and is a dimensionless value. Grounding point With grounding point The mutual resistance correction factor is a dimensionless value. When the electrical distance between two grounding points is greater than 50 meters, the correction factor is 1. For every 10 meters decrease in distance, the correction factor decreases by 0.1 to avoid current dissipation shielding effect between grounding points that are too close together. Grounding point A collection of grounding points that are not within the lightning strike range and have remaining capacity.

[0050] Taking the overloaded grounding point at the ridge location as an example, the normal allocation weight of the corresponding grounding point is 0.13, the maximum distributable current dissipation ratio is 0.09024, the excess current dissipation weight is 0.03976, the thunderstorm arrival time is 4 minutes, the time gradient transfer coefficient is 1, there are 3 grounding points in the vicinity that are not covered by lightning and have remaining capacity, with remaining capacities of 0.04, 0.03, and 0.02 respectively, and mutual resistance correction coefficients of 0.9, 0.8, and 0.6 respectively. Substituting into the formula, the transfer weights received by the three grounding points are calculated to be 0.0192, 0.0142, and 0.00636 respectively. After the transfer is completed, the allocation weight of all grounding points does not exceed their own maximum distributable current dissipation ratio, and there will be no single-point current dissipation overload. The current dissipation weight dynamically adjusted during the thunderstorm period will be used as the basis for temporarily adjusting the conduction status of the controllable current dissipation branch during the thunderstorm. After the thunderstorm has completely passed, it will automatically return to the normal allocation weight.

[0051] After obtaining the final determined current dissipation weights for each grounding point, the corresponding controllable current dissipation branch's conduction quantity and conduction duration control parameters are generated. During the calculation process, a branch lifespan balancing correction term, distinct from the existing proportional conduction method, is added. When selecting controllable current dissipation branches, branches with fewer historical operation counts are prioritized for conduction, balancing the operation frequency of all branches and extending the overall lifespan. The formula for calculating the conduction quantity is as follows: , in the formula For the first The number of controllable current-dissipating branches that need to be connected for each grounding point, expressed in terms of the number of branches, with the result rounded up to the nearest integer. For the first The final current dissipation weight assigned to each grounding point is a dimensionless value. The redundancy coefficient is set to 1.2, reserving 20% ​​redundancy capacity to handle situations where the lightning current amplitude exceeds the prediction or a branch failure occurs. The conduction duration is determined based on the expected duration of the thunderstorm warning period, starting from the moment the thunderstorm warning is triggered and continuing until 15 minutes after the warning is lifted, ensuring sufficient current dissipation capacity to handle subsequent residual lightning after the thunderstorm has completely passed. Using the final weight of 0.09024 for the thunderstorm period at the ridge location grounding point, with a predicted maximum lightning current of 100 kA, a single branch rated current of 2 kA, and a redundancy coefficient of 1.2, the number of branches requiring conduction is calculated as ceil((0.09024×100 / 2)×1.2)=6. The remaining 2 branches are reserved as backups. The 6 branches with the fewest historical action counts are selected for conduction, with the conduction duration from the warning trigger to 15 minutes after the warning is lifted.

[0052] After determining the conduction branch numbers, conduction quantities, conduction times, and turn-off times for all grounding points, all parameters are framed according to the communication protocol to form a current dissipation weight allocation instruction. This instruction is then sent to the grounding hardware module via the field industrial bus. Upon receiving the instruction, the grounding hardware module conducts or turns off the designated controllable current dissipation branches at the corresponding times according to the parameter requirements. During thunderstorm warning periods, while issuing the current dissipation weight allocation instruction, the pre-adjustment results of the current dissipation weight are generated simultaneously. These results include the temporary allocation weight for each grounding point, the total discharge capacity of the grounding network, and the fastest surge discharge time window. The pre-adjustment results of the current dissipation weight are sent to the cable transition lightning protection active control unit for use in subsequent surge prediction calculations. This ensures that the surge prediction process can accurately grasp the actual discharge capacity of the grounding side, avoiding overvoltage at cable junctions caused by mismatched discharge capacity. Compared to the traditional fixed weight allocation method, this dynamic weight allocation method can improve the overall current dissipation capacity utilization rate of the distributed grounding network by more than 40%, while reducing the probability of single-point overload by more than 90%. The branch life balancing strategy can extend the overall lifespan of the controllable current dissipation branch by 30%. Sending the pre-adjustment results to the cable lightning protection side in advance can control the surge discharge coordination error between the grounding side and the cable side to within 1 millisecond, effectively reducing the overvoltage level at the cable connection and comprehensively improving the grounding and lightning protection reliability of mountain distributed wind power projects.

[0053] Existing methods for grounding current dissipation in mountainous wind power projects mostly adopt a fixed ratio allocation mode. This mode does not adjust the allocation ratio based on the dynamic changes in the conductivity of mountainous soil and the spatiotemporal characteristics of thunderstorm movement. This can easily lead to problems such as overload and erosion of grounding points in areas directly hit by thunderstorms, and idle current dissipation capacity of grounding points in areas not directly hit by thunderstorms. This not only makes it difficult to fully utilize the current dissipation potential of the distributed grounding network, but also poses a risk of grounding potential rise and backlash against wind turbines and cable equipment. Based on the actual scenario of the 35MW distributed mountainous wind power project in Pinglu District, this paper implements the entire process of dynamic calculation of current dissipation weight and command generation.

[0054] When calculating the current diffusion weight, basic parameters are first extracted from the mountain current diffusion characteristic map. The extraction process uses the actual burial latitude and longitude coordinates of each grounding point as the matching benchmark. The current diffusion capacity weight of all grids within a 25-meter radius around the grounding point is calculated inversely by distance with the lightning protection level to avoid parameter deviation caused by abnormal data of a single grid. In the weighting process, the weight value of each grid is inversely proportional to the square of the straight-line distance from the grid center to the grounding point. The closer the distance, the higher the weight value. Finally, the regional average current diffusion capacity weight and regional average lightning protection level corresponding to each grounding point are obtained. The regional average current diffusion capacity weight is used for the subsequent calculation of the maximum distributable current diffusion ratio, and the regional average lightning protection level is used for the subsequent determination of current diffusion priority.

[0055] After extracting the basic parameters, the maximum distributable current dissipation ratio for each grounding point is calculated. During the calculation, four dynamic correction terms, distinct from the existing direct conversion method of rated capacity, are added: grounding electrode temperature correction, soil condition correction, slope path correction, and drag-reducing agent aging correction. The formula for calculating the maximum distributable current dissipation ratio is as follows: , in the formula For the first One grounding point at The maximum distributable current distribution ratio at any given time, ranging from 0 to 1, is a dimensionless value. For the first The total number of controllable current-dissipating branches connected to each grounding point, in units of branches. This refers to the inherent rated current capacity of a single controllable current-dissipating branch, expressed in kiloamperes (kA). The inherent rated current capacity is the long-term permissible current capacity determined by type testing at the time of manufacture of the controllable current-dissipating branch. This is the grounding electrode temperature correction factor, which is a dimensionless value. This represents the real-time operating temperature of the grounding electrode at the i-th grounding point, in degrees Celsius. The correction factor is set to 1 at 25 degrees Celsius, and decreases by 0.06 for every 10 degrees Celsius increase in temperature. When a large current flows through the grounding electrode, causing its temperature to rise, the metal resistance of the grounding electrode will increase with the temperature, and its current-carrying capacity will decrease accordingly. This is a soil condition correction factor, which is a dimensionless value. This refers to the real-time soil moisture content around the grounding point. The correction factor is set to 1 when the soil is moist and not frozen, 0.78 when the soil moisture content is below 10%, and 0.65 when the surface soil temperature is below 0 degrees Celsius and freezing occurs. Dry or frozen soil will cause a significant increase in resistivity, reducing current dissipation efficiency. This is the slope path correction factor, which is a dimensionless value. The ground slope at the location of the grounding point is expressed in degrees. The correction factor is 1 when the slope is 0 degrees, and decreases by 0.04 for every 5 degrees increase in slope. As the slope increases, the actual path length of the lightning current along the slope is greater than the horizontal projected length, and the current dissipation resistance increases with the length of the path. This is the aging correction factor for drag-reducing agents, and it is a dimensionless value. The value represents the service life of the grounding point, expressed in years. For newly commissioned grounding points where the resistance-reducing agent is in good working order, the correction factor is set to 1. For each additional year of service life, the correction factor decreases by 0.03. The resistance-reducing agent gradually deteriorates due to rainwater erosion and ion diffusion, leading to an increase in the contact resistance between the grounding electrode and the soil. for The maximum lightning current amplitude predicted at all times based on lightning monitoring data, in kiloamperes.

[0056] Taking the actual calculation of grounding point No. 12 at the ridge location of the project as an example, the corresponding grounding point is connected to a total of 8 controllable current-dissipating branches. The inherent rated current capacity of a single branch is 2 kA. The real-time monitoring grounding electrode temperature is 38 degrees Celsius, with a temperature correction factor of 0.922. The soil moisture content is 8% and there is no freezing after 12 consecutive days of no rainfall, with a soil condition correction factor of 0.78. The ground slope at the location is 27 degrees, with a slope path correction factor of 0.784. The grounding point has been in operation for 3 years, and the resistance-reducing agent aging correction factor is 0.91. The predicted maximum lightning strike... With a current amplitude of 100 kA, the maximum distributable current dissipation ratio calculated by substituting into the formula is (8×2×0.922×0.78×0.784×0.91) / 100≈0.082, which means that the corresponding grounding point can bear a maximum of 8.2% of the total lightning current discharge task. The calculated maximum distributable current dissipation ratio will be used as the hard upper limit of the current dissipation weight allocation. Under any operating condition, the current dissipation weight allocated to a single grounding point shall not exceed the corresponding value to avoid overload of the grounding point, which would cause the ground potential to rise and backfire the fan or cable equipment.

[0057] After calculating the maximum distributable current dissipation ratio, the current dissipation priority of each grounding point is determined. During the calculation, three correction terms, distinct from existing methods that only rank by risk level, are added: a tower base distance correction term, a cable connection point distance correction term, and a branch lifespan balancing correction term. Grounding points closer to the tower base or cable connection point have shorter current dissipation paths, faster current discharge, and higher priority. Grounding points with fewer historical operation counts also have higher priority. Balancing the operation frequency of all branches extends service life. The formula for calculating current dissipation priority is as follows: , in the formula For the first One grounding point at The current propagation priority at any given time; a higher value indicates a higher priority. This value is dimensionless. For the first The average lightning protection level of the area corresponding to each grounding point is an integer ranging from 1 to 4. This is the correction factor for the distance from the base of the tower, and it is a dimensionless value. This represents the electrical distance from the grounding point to the corresponding wind turbine tower base, in meters. A correction factor of 1.2 is used when the distance is 0 meters, and decreases by 0.05 for every 20 meters increase in distance. This is a correction factor for the distance between cable connection points; it is a dimensionless value. This is the electrical distance from the grounding point to the connection point between the corresponding overhead and buried cables, expressed in meters. The correction factor is 1.1 when the distance is 0 meters, and decreases by 0.03 for every 30 meters increase in distance. This is the lifetime balance correction factor, which is a dimensionless value. The average historical number of actions for all controllable current-dissipating branches at the grounding point is 1.1 when the number of actions is 0, and the correction factor decreases by 0.02 for every 10 additional actions.

[0058] Taking the actual calculation of two grounding points within the project as an example, grounding point No. 12 is located next to the wind turbine tower base. The average lightning protection level in the area is level 4. It is 3 meters away from the tower base, with a tower base distance correction factor of 1.1925. It is 18 meters away from the cable connection point, with a cable connection point distance correction factor of 1.082. The average historical number of operations for the branch circuit is 12, and the lifespan balancing correction factor is 1.076. The calculated current dissipation priority is 4 × 1.1925 × 1.082 × 1.076 ≈ 5.56. The grounding point is located below the hillside, with an average lightning protection level of 2 in the area. It is 90 meters away from the tower base, with a tower base distance correction factor of 0.975. It is 160 meters away from the cable junction, with a cable junction distance correction factor of 0.84. The branch's average historical operation count is 3, and the life balance correction factor is 1.094. The calculated current dissipation priority is 2 × 0.975 × 0.84 × 1.094 ≈ 1.79. When allocating current dissipation weights, the current dissipation needs of high-priority grounding points will be prioritized.

[0059] After prioritizing the current dissipation, the current dissipation weight allocation under normal operating conditions is carried out. The allocation process proceeds sequentially from highest to lowest priority. First, the highest priority grounding point is allocated a weight value not exceeding 90% of its maximum dissipable current dissipation ratio, reserving 10% redundancy capacity to cope with sudden faults or lightning current amplitude exceeding the prediction. After the high-priority grounding points are allocated, the weights are then allocated to the next lower priority grounding points until the sum of the allocated weights of all grounding points is 1, meaning that all power frequency ground fault currents or lightning currents have corresponding discharge channels. After the allocation is completed, the allocated weight of each grounding point is checked point by point to see if it exceeds the maximum dissipable current dissipation ratio. If it does, the excess weight is transferred to the next priority grounding point with remaining capacity to ensure that all grounding points operate within the safe capacity range. The current dissipation weights allocated under normal operating conditions will serve as the basis for controlling the number of controllable current dissipation branches during daily operation. The corresponding number of branches are kept in operation daily to meet the discharge requirements of power frequency ground faults and small-amplitude lightning currents.

[0060] During thunderstorm warning periods, the initially allocated current dissipation weights are dynamically adjusted based on lightning path data. The adjustment process first identifies grounding points within the lightning path's coverage area. The arrival time of the thunderstorm at the corresponding grounding point is calculated based on the lightning's speed. A gradient transfer coefficient is set according to the arrival time: 0.3 for arrival times greater than 10 minutes, 0.6 for arrival times between 5 and 10 minutes, and 1 for arrival times less than 5 minutes. The difference between the current allocated weight and the maximum dissipable current dissipation ratio for each grounding point is calculated. A negative difference indicates an overload risk at the grounding point during a direct thunderstorm strike, requiring the transfer of excess current dissipation weight to surrounding grounding points. Two correction terms, distinct from the existing nearest-neighbor transfer, are added during the weight transfer process: a mutual resistance shielding correction term and a soil conductivity correction term. The closer the electrical distance between two grounding points, the lower the mutual resistance, resulting in a stronger current shielding effect during current dissipation and a lower transferable weight ratio. Stronger soil conductivity around the target grounding point allows for a higher acceptable transfer weight. The formula for calculating the weight transfer amount is as follows: , in the formula overload grounding point It needs to be transferred to the surrounding grounding point. The divergence weight is a dimensionless value. Grounding point Excess current distribution weights exceeding the maximum allocatable current distribution ratio are dimensionless values. Here, represents the time gradient transfer coefficient, which is a dimensionless value. Grounding point The remaining allocable capacity is the difference between the maximum allocable current distribution ratio and the current allocation weight, and is a dimensionless value. Grounding point With grounding point The mutual resistance shielding correction factor is a dimensionless value. When the electrical distance between two grounding points is greater than 50 meters, the correction factor is 1. For every 10 meters decrease in distance, the correction factor decreases by 0.1 to avoid current dissipation shielding effects between grounding points that are too close, which would reduce the actual discharge capacity. Grounding point The correction factor for the conductivity of the surrounding soil is a dimensionless value. The correction factor is 1.1 when the soil is moist and 0.9 when the soil is dry. S is the grounding point. A collection of grounding points that are not within the lightning strike range and have remaining capacity.

[0061] Taking the No. 12 overload grounding point at the ridge location as an example, the normal allocation weight of the corresponding grounding point is 0.135, the maximum distributable current dissipation ratio is 0.082, the excess current dissipation weight is 0.053, the thunderstorm arrival time is 4 minutes, the time gradient transfer coefficient is 1, there are 3 grounding points in the vicinity that are not covered by lightning and have remaining capacity, with remaining capacities of 0.05, 0.03, and 0.02 respectively, mutual resistance shielding correction coefficients of 0.9, 0.8, and 0.6 respectively, and soil conductivity correction coefficients of 1.1, 1.0, and 0.9 respectively. Substituting into the formula, the transfer weights received by the three grounding points are 0.0263, 0.0143, and 0.0124 respectively. After the transfer is completed, the allocation weight of all grounding points does not exceed their own maximum distributable current dissipation ratio, and there will be no single-point current dissipation overload. The current dissipation weight dynamically adjusted during the thunderstorm period will be used as the basis for temporarily adjusting the conduction status of the controllable current dissipation branch during the thunderstorm. After the thunderstorm has completely passed, it will automatically return to the normal allocation weight.

[0062] After obtaining the final determined current dissipation weights for each grounding point, the corresponding controllable current dissipation branch's conduction quantity and conduction duration control parameters are generated. During the calculation process, a branch life balancing strategy, different from the existing proportional conduction strategy, is incorporated. When selecting controllable current dissipation branches, branches with fewer historical operation counts are prioritized for conduction, balancing the operation frequency of all branches and extending the overall service life of the grounding grid. The formula for calculating the conduction quantity is as follows: , in the formula For the first The number of controllable current-dissipating branches that need to be connected for each grounding point, expressed in terms of the number of branches, with the result rounded up to the nearest integer. For the first The final current dissipation weight assigned to each grounding point is a dimensionless value. The redundancy coefficient is set to 1.2, reserving 20% ​​of the redundancy capacity to cope with situations where the lightning current amplitude exceeds the prediction or a single branch fault occurs. The conduction time is determined based on the expected duration of the thunderstorm warning period, starting from the moment the thunderstorm warning is triggered and continuing until 15 minutes after the thunderstorm warning is lifted, ensuring that there is still sufficient current dissipation capacity to cope with subsequent residual lightning after the thunderstorm has completely passed.

[0063] Based on the final weight of 0.082 for the thunderstorm period at grounding point No. 12, the predicted maximum lightning current is 100 kA, the rated current capacity of a single branch is 2 kA, and the redundancy coefficient is 1.2. Substituting these values ​​into the formula, the number of branches that need to be connected is calculated to be ceil((0.082×100 / 2)×1.2)=5. The remaining 3 branches are reserved as backups. The 5 branches with the fewest historical action counts are selected for connection. The connection duration is from the triggering of the warning to 15 minutes after the warning is lifted.

[0064] After determining the conduction branch numbers, conduction quantities, conduction times, and turn-off times for all grounding points, all parameters are framed according to the field industrial bus communication protocol to form a current dissipation weight allocation instruction. This instruction is then sent to the grounding hardware module via a shielded communication cable. Upon receiving the instruction, the grounding hardware module conducts or turns off the designated controllable current dissipation branches at the corresponding times according to the parameter requirements. During thunderstorm warning periods, while issuing the current dissipation weight allocation instruction, the pre-adjustment result of the current dissipation weight is generated simultaneously. This result includes the temporary allocation weight for each grounding point, the total discharge capacity of the grounding network, and the fastest surge discharge time window. The pre-adjustment result of the current dissipation weight is sent to the cable transition lightning protection active control unit for use in subsequent surge prediction calculations. This ensures that the surge prediction process can accurately grasp the actual discharge capacity of the grounding side and avoid overvoltage at cable connections caused by mismatched discharge capacity.

[0065] Compared to the traditional fixed weight allocation method, this dynamic weight allocation method can improve the overall current dissipation capacity utilization rate of the distributed grounding network by more than 42%, while reducing the probability of single-point overload by more than 93%. The branch life balancing strategy can extend the overall service life of the controllable current dissipation branch by 35%. Sending the pre-adjustment results to the cable lightning protection side in advance can control the surge discharge coordination error between the grounding side and the cable side to within 1 millisecond, effectively reducing the overvoltage level at the cable connection and comprehensively improving the grounding and lightning protection reliability of mountain distributed wind power projects.

[0066] Existing grounding branch adjustments often employ a one-time full-open / full-close action mode, failing to consider the dynamic changes in the actual current-carrying capacity of the branch, and neglecting to set action intervals to suppress operational overvoltage. During thunderstorms, the fixed redundancy quantity easily leads to insufficient redundancy or ineffective branch actions, which can easily cause operational overvoltage impacts on equipment and accelerate the wear of branch switches. Based on the actual scenario of the 35MW distributed mountain wind power project in Pinglu District, a full-process implementation of controllable current-dissipating branch conduction status adjustment was carried out. After receiving the current-dissipating weight allocation command, the grounding hardware module first verifies the absence of bit flips during command transmission using a 16-bit CRC check to avoid malfunctions caused by communication errors. After successful verification, it parses the conduction quantity and conduction duration control parameters corresponding to each grounding point carried in the command. During parsing, the legality range of each parameter is checked. The conduction quantity cannot exceed the total number of controllable current-dissipating branches connected to the corresponding grounding point, and the conduction duration cannot be less than 0 or exceed 24 hours. If the parameters exceed the legal range, a parameter error flag is sent back requesting a resend command. After the parameter verification is successful, the actual current-carrying capacity comparison stage begins.

[0067] The comparison process does not directly use the rated current capacity of the branch for simple difference calculation. Instead, it incorporates three dynamic correction terms, distinct from existing rated value comparisons: a branch contact resistance correction term, a branch temperature correction term, and a current dissipation efficiency correction term for branches at different burial depths. This calculates the matching difference between the actual available current capacity and the required current capacity of the currently conductive branch at each grounding point. The formula for calculating the matching difference is as follows: , in the formula For the first One grounding point at The current capacity matching difference at any given time, in kiloamperes. A positive difference indicates that additional conducting branches are needed to improve current dissipation capability, while a negative difference indicates that some branches need to be shut off to reduce ineffective conduction. The total current capacity required for the grounding point, in kiloamperes. The final current dissipation weight assigned to the grounding point is a dimensionless value. To predict the maximum lightning current amplitude, the unit is kiloampere. This represents the actual total available current capacity of the currently conductive branches at the grounding point, in kiloamperes. This represents the number of controllable current-dissipating branches currently connected at the grounding point, expressed in units of branches. This refers to the inherent rated current capacity of a single controllable current-dissipating branch, in kiloamperes. For the first The contact resistance correction factor for an already conductive branch is a dimensionless value. For the first The real-time contact resistance of the branch circuit breaker is measured in milliohms. When the contact resistance is 0, the correction factor is 1. For every 1 milliohm increase in contact resistance, the correction factor decreases by 0.05. Excessive contact resistance can lead to severe heating during branch circuit current flow, resulting in a decrease in actual current-carrying capacity. For the first The temperature correction factor for each branch is a dimensionless value. For the first The real-time temperature of the branch circuit switch is in degrees Celsius. The correction factor is 1 when the temperature is 25 degrees Celsius, and decreases by 0.06 for every 10 degrees Celsius increase in temperature. For the first The burial depth-based flow dispersion efficiency correction coefficient for this branch road is a dimensionless value. The burial depth of the branch road is in meters. The correction factor is 1 when the burial depth is 2 meters. The correction factor decreases by 0.03 for every 0.5 meters decrease in burial depth. The surface soil resistivity fluctuates greatly, and the flow dispersion stability of shallow-buried branch roads is weaker than that of deep-buried branch roads.

[0068] Taking the actual calculation of the grounding point on ridge No. 12 as an example, the current dissipation weight allocated to the corresponding grounding point is 0.082. The predicted maximum lightning current amplitude is 100 kA, and the required total current carrying capacity is 8.2 kA. Currently, three controllable current dissipation branches are connected, with an inherent rated current carrying capacity of 2 kA for each branch. The average contact resistance of the three branches is 1 milliohm, with a contact resistance correction factor of 0.95. The average switch temperature is 38 degrees Celsius, with a temperature correction factor of 0.922. The average burial depth of the branches is 1 meter. The depth of the burial is 0.5 meters, and the current dissipation efficiency correction coefficient is 0.97. The actual usable total current carrying capacity of the already connected branch is 3×2×0.95×0.922×0.97≈5.09 kA. The calculated matching difference is 8.2-5.09≈3.11 kA, which is a positive value. This indicates that the grounding point needs to increase its current dissipation capacity. The calculated matching difference will serve as the direct basis for adding or closing the branch. The sign of the difference determines the direction of adjustment, and the magnitude of the difference determines the number of branches that need to be operated.

[0069] After determining the adjustment direction, the controllable current-dissipating branches to be operated are prioritized. The prioritization process incorporates four correction factors distinct from the existing random selection: branch historical operation count correction, branch burial depth adaptation correction, branch contact resistance correction, and branch fault probability correction. Branches with higher priority are turned on first, while branches with lower priority are turned off first. The formula for calculating branch operation priority is as follows: , in the formula For the first The first grounding point The action priority of each branch is given by a value; higher values ​​indicate higher priority for being turned on, while lower values ​​indicate higher priority for being turned off. These values ​​are dimensionless. The distribution capacity weight of the grid where the branch is located is a dimensionless value. This is a correction factor for the number of historical actions, and it is a dimensionless value. The historical cumulative number of actions for each branch is used. When the number of actions is 0, the correction coefficient is 1.1. For every 10 additional actions, the correction coefficient decreases by 0.02. Branches with fewer actions are prioritized for activation to balance the lifespan of all branches. A correction factor, dimensionless, is used to adapt to different burial depths. When the surface soil is dry or frozen, deeper burial branches are preferred, and the correction factor increases with increasing burial depth. When the surface soil is moist, shallower burial branches are preferred, and the correction factor decreases with increasing burial depth. This is a contact resistance correction factor, a dimensionless value. The smaller the contact resistance, the higher the correction factor. Branches with good contact conditions are preferentially selected for conduction. This is a fault probability correction coefficient, a dimensionless value, calculated based on the branch's historical fault records. The correction coefficient is 1 when there are no fault records, and 0.5 for branches that have experienced failure to operate or malfunction, thus reducing the priority of the faulty branches. Taking the five unconnected branches awaiting connection at grounding point 12 as an example, the priority is calculated based on each branch's current dissipation capacity weight, historical operation count, real-time contact resistance, and burial depth. These branches are then sorted from highest to lowest priority, and the sorting result will serve as the basis for selecting the branch operation sequence.

[0070] The adjustment process adopts a step-by-step differential action mode, unlike the existing method of simultaneous all-around action. Each step only operates one branch, with a 100-millisecond time interval between adjacent actions. This avoids operational overvoltage impacts on cables or fan equipment caused by simultaneous operation of multiple branches. For grounding points with positive matching differences requiring increased current dissipation capacity, the controllable current dissipation branches that are not yet in operation are sequentially activated according to the priority of the branches to be activated from high to low. After activating one branch each time, the current actual available current capacity is recalculated until the actual total available current capacity is greater than or equal to the required current capacity. Before activating a branch, the following steps are performed: The insulation resistance of the branch is detected. If the insulation resistance is lower than the threshold, it is determined to be an insulation fault in the branch. The branch is automatically skipped, and the next priority branch is selected for conduction. For grounding points with negative matching difference that require reduced current dissipation capacity, the controllable current dissipation branches that are already conducting are turned off one by one in order of priority from low to high. After each branch is turned off, the current actual available current capacity is recalculated until the actual total available current capacity matches the required current capacity. During the turn-off, branches with a high number of historical actions, high contact resistance, and poor current dissipation stability are given priority to reduce the ineffective action time of high-performance branches. In addition to the above differential adjustment actions, the pre-adjustment during thunderstorm warning periods dynamically calculates the number of redundant conducting branches based on the received pre-adjustment results of the current dissipation weight. Unlike the existing fixed redundancy method, the redundancy number is dynamically adjusted according to the lightning current amplitude margin and the lightning protection level of the grounding point. The formula for calculating the number of redundant branches is as follows: , in the formula This represents the number of redundant branches that require additional conduction, expressed in units of branches. The result is rounded up to the nearest integer. This is a correction factor for lightning protection level, a dimensionless value. The correction factor is 1.1 for protection level 1, and increases by 0.1 for each additional protection level. Higher protection levels result in a larger redundancy ratio. is the average correction coefficient for the conducted branches, and is a dimensionless value.

[0071] For the grounding point receiving the transferred current dissipation weight, in addition to the required number of branches, an extra number of controllable current dissipation branches with calculated redundancy are turned on to reserve current dissipation redundancy capacity. In case of lightning current amplitude exceeding the prediction or single branch failure, for the grounding point transferring the current dissipation weight, the number of controllable current dissipation branches corresponding to the current dissipation weight requirement are kept on, and excess branches are turned off to reduce unnecessary power loss.

[0072] Taking grounding point No. 12 as an example, the lightning protection level of the corresponding grounding point is level 4, the protection level correction factor is 1.4, the average correction factor of the already connected branches is 0.95, and the number of redundant branches is calculated to be ceil((0.082×100×(1.4-1)) / (2×0.95))=2. On the basis of the requirement of 5 branches, 2 additional redundant branches are connected to reserve sufficient current dissipation margin.

[0073] After all controllable current-dissipating branches are adjusted, the actual conduction status of each grounding point is verified point by point. The verification process is not limited to switch position feedback. The loop resistance of the conducted branches is detected one by one. When the loop resistance exceeds the threshold, it is determined to be a branch conduction fault. The next priority backup branch is automatically selected to replace the faulty branch. This process continues until the loop resistance of all branches meets the current carrying requirements and the actual available current carrying capacity of all grounding points meets the current dissipation weight requirements. Then, a conduction status adjustment completion identifier is generated. The identifier includes the actual number of conducted branches, the actual total available current carrying capacity, the faulty branch number, and the adjustment completion time for each grounding point. The conduction status adjustment completion identifier is uploaded to the multi-node status fusion management and control unit as part of the full node operation data for subsequent anomaly identification and model iteration. This differential step adjustment method with dynamic correction can reduce the amplitude of operating overvoltage caused by branch operation by more than 60%. Multi-dimensional priority sorting can extend the overall service life of controllable current dissipation branches by 40%. The automatic replacement logic for faulty branches can improve the success rate of current dissipation adjustment to 99.9%. Dynamic redundancy configuration can reduce daily branch current loss while ensuring current dissipation margin during thunderstorms, comprehensively improving the reliability and safety of the grounding adjustment process.

[0074] Conventional thunderstorm warnings often use a fixed electric field threshold and average moving speed extrapolation method, without considering the distortion effect of mountainous terrain on the atmospheric electric field and the acceleration changes of thunderstorm movement. They also fail to combine the reliability differences between local high-precision monitoring data and regional lightning location data for fusion calculation, which easily leads to false alarms, missed alarms, or insufficient warning time. This makes it difficult to gain enough preparation time for grounding pre-adjustment and active lightning protection. Based on the actual scenario of the 35MW mountainous distributed wind power project in Pinglu District, we carried out the entire process of thunderstorm warning time determination.

[0075] Real-time access to local electric field strength, lightning proximity signals, lightning movement paths, lightning strike locations, and lightning current amplitude data is used. This data, combined with a regional lightning risk probability distribution layer, is used for tiered assessment. The lightning current amplitude data is composed of direct lightning current monitoring values ​​collected by traveling wave monitoring devices deployed locally at the wind farm and cloud-to-ground lightning current monitoring values ​​transmitted back from the regional lightning location system. The lightning current amplitude data from these two sources are weighted and fused based on source reliability. The formula for calculating the fused lightning current amplitude is as follows: , in the formula The amplitude of the merged lightning current is expressed in kiloamperes. The weight of local traveling wave monitoring data is set to 0.7. Local traveling wave devices, being closer to the wind field and having higher measurement accuracy, receive a higher weight. The amplitude of direct lightning strike current collected by the local traveling wave monitoring device, in kiloamperes. The weight for the regional lightning location system data is set to 0.3. The regional location system has a wide coverage area but lower accuracy due to long distances, so the weight is lower. The amplitude of cloud-to-ground lightning current transmitted back by the regional lightning location system is in kiloamperes. When the local traveling wave device does not detect lightning current, only regional location data is used. When the regional location system has no data, only local monitoring data is used. This avoids judgment failure caused by the failure of a single data source. The fused lightning current amplitude will be used for lightning movement speed calculation and lightning intensity determination.

[0076] Before conducting initial early warning assessments, a terrain electric field correction coefficient is added. This differs from existing fixed-threshold assessment methods. The electric field threshold is corrected based on the terrain protrusion height at the electric field meter's installation location, avoiding false alarms caused by electric field distortion at ridge locations. The formula for calculating the corrected electric field threshold is as follows: , in the formula The primary warning electric field threshold is calculated after terrain correction, in kilovolts per meter. The initial warning electric field threshold for flat terrain is set at 6 kV per meter. This is the terrain correction factor, which is a dimensionless value. The threshold is the height of the electric field meter installation location relative to the average terrain within a 100-meter radius, expressed in meters. When the height is 0, the correction factor is 1. For every 10 meters increase in height, the correction factor decreases by 0.05. The electric field at prominent locations such as ridges will be enhanced due to terrain distortion, so the threshold should be appropriately increased to avoid false alarms.

[0077] When the real-time monitored local electric field strength reaches the corrected warning threshold, and a lightning proximity signal indicates that the thunderstorm cloud has entered a preset range of 20 kilometers around the wind farm, a primary thunderstorm warning sign is generated. After the primary thunderstorm warning sign is generated, no on-site pre-adjustment actions are triggered. Instead, the data acquisition frequency is increased, shortening the lightning data acquisition cycle from 1 minute to 10 seconds, and the warning preparation state is entered in advance.

[0078] When conducting formal thunderstorm warning assessments, the first step is to calculate the thunderstorm's moving speed and acceleration. Unlike existing methods that only calculate average speed, this approach uses the lightning strike locations at three consecutive moments to calculate the average speed and acceleration of the thunderstorm's movement. The formula for calculating the thunderstorm's moving speed is as follows: , in the formula for The speed of thunderstorm movement at any given moment, measured in meters per second. for Location of lightning strike at all times and Location of lightning strike at all times The straight-line distance between them, in meters. The time difference between two lightning strikes is expressed in seconds. The acceleration of a thunderstorm is the ratio of the velocity difference between two consecutive time intervals to the time difference, expressed in meters per second squared.

[0079] The estimated time for the thunderstorm to reach the wind farm boundary is calculated based on its current location, moving speed, and acceleration. Taking into account the effect of acceleration, the formula for calculating the estimated time is as follows: , in the formula The estimated time, in seconds, for the thunderstorm to reach the boundary of the wind farm. This represents the acceleration of thunderstorm movement, measured in meters per second squared. The distance from the current location of the thunderstorm to the boundary of the wind farm is expressed in meters. When the acceleration is 0, the estimated duration is obtained by dividing the distance by the velocity.

[0080] When lightning path data shows that the direction of the thunderstorm is pointing towards the wind farm area, and the calculated estimated time for the thunderstorm to reach the wind farm boundary is less than 10 minutes of warning preparation time, and at the same time the grounding protection range marked by the lightning risk probability distribution layer of the area covered by the thunderstorm path is determined, the thunderstorm warning period is officially entered. In the determination process, the weight correction of the regional risk layer is added. If the lightning protection level of the grid covered by the thunderstorm path is level 4, the warning preparation time is extended to 15 minutes to give more preparation time to high-risk areas.

[0081] Once the thunderstorm warning period officially begins, the grounding point current dissipation collaborative optimization unit synchronously triggers the pre-adjustment result of the current dissipation weight to be sent to the cable transition lightning protection active control unit, triggering the pre-adjustment action of the field controllable current dissipation branch.

[0082] Taking a specific thunderstorm event during the project as an example, the local monitoring showed that the electric field strength reached the corrected threshold of 6.8 kV / m. A lightning proximity signal indicated that the thunderstorm cloud was located 18 kilometers around the wind farm, generating a primary warning. Subsequently, three consecutive lightning strikes were detected. The calculated speed of the thunderstorm was 120 km / h, with an acceleration of 5 km / h squared. The current location of the thunderstorm was 18 kilometers from the boundary of the wind farm, with an estimated arrival time of 8.7 minutes. The thunderstorm path covered a ridge area with a level 4 protection rating, thus officially entering the thunderstorm warning period and triggering pre-adjustment actions.

[0083] The method for determining the cancellation of early warnings differs from the existing single-condition cancellation method. It adopts a continuous dual-cycle verification mechanism to avoid missed reports caused by thunderstorm recurrence or southerly winds. When no new lightning strike location is collected within two consecutive collection cycles (20 seconds), and the local electric field strength drops below the safe threshold of 3 kV / m, and the electric field strength shows a downward trend for two consecutive cycles, the thunderstorm warning period is determined to be over, the pre-adjustment state is cancelled, and the controllable current-dissipating branches are restored to the normal conduction quantity. If a new lightning strike occurs or the electric field strength rises again during the verification process, the verification time is reset, and the warning state is re-determined.

[0084] This tiered early warning system with terrain correction and acceleration extrapolation can reduce the false alarm rate of thunderstorm warnings by more than 60%, and the average warning lead time can reach more than 12 minutes. The dual-cycle cancellation mechanism can reduce the missed alarm rate of thunderstorms by more than 90%, providing sufficient preparation time for grounding pre-adjustment and active lightning protection actions, and avoiding untimely protection actions caused by delayed warnings.

[0085] Conventional cable surge triggering often uses a fixed amplitude threshold triggering method, which does not consider the frequency-related attenuation characteristics of lightning current transmission along the cable, nor does it combine the dynamic current dissipation capability of the grounding side and waveform characteristics to distinguish interference signals. This can easily lead to false triggering or untimely triggering, making it difficult to achieve accurate early conduction before the surge arrives. Based on the actual scenario of the 35MW mountain distributed wind power project in Pinglu District, we will carry out the full-process implementation of cable surge prediction and triggering.

[0086] The pre-adjustment results of the current dissipation weight during the thunderstorm warning period are used to summarize the current dissipation weights of all grounding points within a 100-meter radius of the connection point between overhead and buried cables. Combining the actual current carrying capacity of the controllable current dissipation branch connected to each grounding point with the cable wave refraction coefficient, the total current dissipation capacity around the connection point is calculated. The formula for calculating the total current dissipation capacity is as follows: , in the formula The total current dissipation capacity around the connection location is expressed in kiloamperes (kA). This is the refraction coefficient of lightning current as it travels from overhead cable to buried cable. It is a dimensionless value, calculated based on the wave impedance of the overhead and buried cables, with a typical value of 0.8. This refers to the set of grounding points within a 100-meter radius of the connection point. For the first The distributed current dissipation weights for each grounding point are dimensionless values. For the first The number of controllable current-dissipating branches that are connected at each grounding point, in units of branches. The inherent rated current carrying capacity of a single controllable current dissipation branch is expressed in kiloamperes. The calculated total current dissipation capacity will be used as a benchmark value for comparing surge amplitudes to determine whether an additional discharge path is needed.

[0087] After completing the total current dissipation capacity calculation, the traveling wave monitoring device installed at the junction of the overhead cable and the buried cable collects the cable traveling wave signal, real-time local electric field strength, real-time lightning current amplitude, and lightning movement path data. Based on the lightning movement path data and the expected arrival time of the thunderstorm, the time window for the surge to reach the junction is initially calculated. The starting point of the time window is the expected arrival time of the thunderstorm minus the inherent conduction time of the controllable discharge path, and the ending point is the expected arrival time of the thunderstorm plus 2 minutes of redundancy time. The calculated time window is used to control the sensitivity of traveling wave identification. Outside the time window, the traveling wave trigger threshold is increased by 3 times to reduce interference and false triggering, while within the time window, a high-sensitivity identification state is maintained.

[0088] When identifying traveling wave signals, a dual-time-domain feature determination method, distinct from existing fixed-amplitude identification, is incorporated. The rise time of the wavefront and the half-peak time of the wave tail are calculated separately. For lightning surges, the rise time of the wavefront is less than 1.2 microseconds, and the half-peak time of the wave tail is less than 50 microseconds. For operational overvoltages or other interference signals, the rise time of the wavefront is greater than 5 microseconds, and the half-peak time of the wave tail is greater than 200 microseconds. Only when both the rise time of the wavefront and the half-peak time of the wave tail simultaneously meet the characteristics of a lightning surge are they considered valid lightning surge signals, thus avoiding false triggering caused by switching operations or partial discharges. After identifying a valid lightning surge waveform, the actual amplitude of the surge reaching the connection point is calculated by combining the real-time lightning current amplitude with a dynamically calculated equivalent cable transmission attenuation coefficient. Unlike existing fixed attenuation coefficient calculation methods, the equivalent attenuation coefficient is dynamically determined based on the dominant frequency of the lightning current. The higher the lightning current frequency, the more pronounced the skin effect and the greater the cable transmission attenuation. The formula for calculating the actual surge amplitude is as follows: , in the formula The actual surge amplitude at the connection point, in kiloamperes. The amplitude of the lightning current after dual-source fusion is expressed in kiloamperes. This is the frequency-dependent equivalent cable transmission attenuation coefficient, measured in netops per meter. The attenuation coefficient increases with increasing dominant frequency of the lightning current. The length of the cable from the lightning strike point to the connection point is in meters. The cable transmission attenuation coefficient no longer uses a fixed calibration value, but is dynamically calculated as an equivalent value based on the main frequency of the lightning current for each lightning strike, which greatly improves the accuracy of amplitude calculation.

[0089] Taking a specific lightning strike in a project as an example, the lightning strike point is 2 kilometers away from the connection point. After merging, the lightning current amplitude is 50 kA, the main frequency of the lightning current is 100 kHz, the corresponding equivalent attenuation coefficient is 0.0002 neppers per meter, and the wave refraction coefficient is taken as 0.8. Substituting these values ​​into the formula, the actual surge amplitude reaching the connection point is calculated to be 50 × 0.8 × exp(-0.0002 × 2000) ≈ 26.8 kA. The calculated actual surge amplitude will be used for comparison of discharge capacity. The actual surge amplitude will be compared with the total current dissipation capacity around the connection point. When the actual amplitude exceeds the rated discharge capacity of the normally conductive discharge path in the cable lightning protection module, the number of additional controllable discharge paths required will be calculated based on the difference between the actual amplitude and the rated capacity of the normally conductive discharge path. A 20% redundancy factor will be added to the calculation to handle situations where the lightning current amplitude exceeds expectations. The formula for calculating the number of additional conductive paths is as follows: , in the formula This represents the number of additional controllable discharge paths that need to be connected, expressed in units of [number] paths. The result is rounded up to the nearest integer. This represents the total rated discharge capacity of the normally open discharge path, expressed in kiloamperes (kA). This refers to the rated discharge capacity of a single controllable discharge path, expressed in kiloamperes (kA). With a redundancy factor of 1.2, when the actual amplitude is less than or equal to the rated discharge capacity of the normally open discharge path, the number of additional conduction paths is 0, and no additional conduction action is triggered.

[0090] After calculating the number of additional conductive paths required, the paths are sorted from least to most historically active. Pathways with fewer historical active paths are prioritized for activation to balance the lifespan of all paths. A trigger command is generated and output to the cable surge protection module. The command includes the path number to be activated and the precise activation time to ensure that all paths are fully activated when the surge reaches the connection point. After the trigger command is issued, the activation status of each path is checked. If a path fails to activate, the next priority backup path is automatically activated to supplement the discharge capacity. At the same time, the trigger time, the number of activated paths, and the actual surge amplitude are recorded to generate the trigger execution status, which is then sent back to the mountain grounding characteristic dynamic modeling unit as the basis for dynamic updating of the mountain current dissipation characteristic map.

[0091] This surge prediction method, which combines frequency-dependent attenuation calculation and dual-feature waveform recognition, can reduce the false alarm rate of surge triggering by more than 80%, control the surge amplitude calculation error to within 5%, and ensure that the coordination error of early conduction is less than 1 millisecond. It can not only avoid the loss of path life caused by false triggering, but also ensure that the discharge path is fully conductive when the surge arrives, effectively reducing the overvoltage level at the connection point.

[0092] Conventional multi-node status management often adopts a single threshold over-limit alarm mode, which does not consider the time synchronization error of multi-source data and sensor measurement error, and is prone to false alarms or missed alarms. The extracted correction parameters are not smoothed, which can lead to fluctuations in the diffused flow model and make it difficult to provide reliable data support for the dynamic updating of the diffused flow characteristic map of the mountain. Based on the actual scenario of the 35MW mountain distributed wind power project in Pinglu District, we carried out the full-process implementation of multi-node status fusion management and parameter feedback.

[0093] When conducting status control, three types of operational data are first accessed. The first type is the conduction status adjustment completion flag uploaded by the grounding hardware module, along with the actual conduction status of each controllable current-dissipating branch and real-time grounding resistance data for each grounding point. The second type is the trigger execution status uploaded by the cable lightning protection module, along with the actual conduction status and conduction resistance data for each controllable discharge path. The third type is the real-time soil resistivity data and lightning activity monitoring data uploaded by the field sensing module. The real-time grounding resistance data is the grounding resistance value collected by the field sensing module at each grounding point. After accessing the multi-source data, timestamp alignment is first performed. A unified GPS time reference is used to align data from different sampling frequencies to a 100-millisecond time slice. The formula for calculating the time offset is... , in the formula For the first The time offset of each data item, in milliseconds. The sampling timestamp is provided by the data itself. As the base time for the corresponding time slice, data with a time offset of less than 50 milliseconds is considered valid data and participates in subsequent calculations, while data with a time offset greater than 50 milliseconds is marked as invalid data and discarded directly to avoid misjudgment of status caused by asynchronous data. The aligned valid data will serve as the basic data source for anomaly detection.

[0094] After data alignment is completed, the various actual operating data and the corresponding command requirements are compared point by point with the preset thresholds. Anomaly confidence calculation is added during the comparison process. Unlike the existing mode of alarming when the limit is exceeded in a single cycle, anomalies are only determined to be valid when multiple consecutive sampling cycles meet the abnormal conditions and the cross-validation of multi-source data is consistent. This avoids false alarms caused by sensor transient errors or communication errors. Anomaly determination is divided into two categories. One category is grounding-side anomalies, which include three specific situations. The first is that the actual conduction state of the controllable current dissipation branch does not match the current dissipation weight allocation command requirements. Anomalies are determined when the branch required to be on by the command is in an off state for two consecutive sampling cycles, or the branch required to be off by the command is in a on state for two consecutive sampling cycles. Inconsistent states in a single sampling cycle are determined to be an anomaly due to the transient failure of the switch action to trigger the anomaly. The second is that the real-time grounding resistance of the grounding point exceeds the threshold, and the grounding resistance exceeds the threshold for three consecutive sampling cycles. Anomalies are defined as follows: resistance greater than 10 ohms and no corresponding reasonable change in soil resistivity at the same location; resistance exceeding the limit in a single sampling cycle is considered measurement interference and does not trigger anomalies; the third type is when the real-time soil resistivity change exceeds the threshold, with a soil resistivity change exceeding 50% within 10 minutes and no corresponding rainfall or temperature change data to support it; resistivity fluctuations caused by normal meteorological changes are not considered abnormal. Another type is lightning protection-side anomalies, including three specific situations: the first is when the actual conduction state of the controllable discharge path does not match the trigger command requirements, and the state is inconsistent for two consecutive sampling cycles; the second is when the conduction resistance exceeds the threshold, and the conduction resistance is greater than 10 milliohms for two consecutive sampling cycles after the controllable discharge path is turned on; the third is when the controllable discharge path fails to reset properly after a thunderstorm, and the controllable discharge path remains in a conducting state 15 minutes after the thunderstorm warning is lifted.

[0095] After a valid abnormal state is identified, alarm information of corresponding level is generated according to the severity of the abnormality. The alarm information includes the location of the abnormal point and the type of abnormality. Severe abnormalities include failure of controllable current dissipation branch to operate, failure of controllable discharge path to operate, and grounding resistance exceeding twice the threshold. After the alarm is generated, it is immediately pushed to the corresponding operation and maintenance alarm terminal. The operation and maintenance alarm terminal is a terminal device used to receive abnormal alarm information. General abnormalities include slight grounding resistance exceeding the limit, discharge path reset delay, and small fluctuation of soil resistivity. After the alarm is generated, it is recorded in the local log and summarized and pushed once an hour to reduce the interference of frequent alarms to operation and maintenance personnel.

[0096] After the alarm is pushed, corresponding correction parameters are extracted based on the identified anomaly type. Correction weights are calculated during the extraction process; the more severe the anomaly, the higher the correction weight. Data is fused according to these weights during model updates to avoid significant model fluctuations caused by single measurement errors. When the anomaly is a grounding resistance anomaly caused by changes in soil resistivity, the changed soil resistivity value and the corresponding point coordinates are extracted as correction parameters. The larger the resistivity change, the higher the correction weight; a change of 100% results in a correction weight of 1, and a change of 30% results in a correction weight of 0.3. When the anomaly is a conduction anomaly caused by a fault in a controllable current-dissipating branch or controllable discharge path, the location of the faulty branch or path and the remaining actual current-dissipating capacity after the fault are extracted as correction parameters. The remaining actual current-dissipating capacity is calculated based on the inherent rated conduction capacity of the remaining conductive controllable current-dissipating branch or controllable discharge path after the fault. The formula for calculating the remaining actual current-dissipating capacity is as follows: , in the formula This represents the actual remaining current dissipation capacity after a fault, in kiloamperes (kA). This refers to the number of branches or pathways that remain normally conductive after a fault, expressed in units of [number of branches or pathways]. This refers to the inherent rated current capacity of a single branch or circuit, expressed in kiloamperes (kA). The average correction coefficient for the remaining branches or paths is a dimensionless value that includes temperature correction, contact resistance correction, and burial depth correction. When the anomaly is that the deviation between the actual lightning strike location and the predicted location of the lightning movement path exceeds the threshold, the actual lightning strike location and the actual lightning current amplitude are extracted as correction parameters. The larger the location deviation, the higher the correction weight.

[0097] Taking two actual anomalies in the project as examples, the first was an anomaly at the No. 12 ridge grounding point. The grounding resistance measured for three consecutive sampling cycles was 12 ohms, exceeding the 10-ohm threshold. The soil resistivity at the same location changed from 200 ohm-meters to 260 ohm-meters within 10 minutes, a change of 30%. There was no corresponding rainfall or temperature change data, so it was determined to be an anomaly in the grounding resistance on the grounding side. The soil resistivity of 260 ohm-meters after the change and the corresponding point coordinates were extracted as correction parameters, with a correction weight of 0.3. The second was an anomaly in the No. 3 controllable discharge path after a thunderstorm. The No. 3 path remained conductive 15 minutes after the thunderstorm warning was lifted, which was determined to be an anomaly in the lightning protection side discharge path not being reset. The location of the faulty path was extracted. After the fault, there were 9 remaining normal conductive paths. The rated current capacity of a single path is 2 kA. The average correction coefficient of the remaining paths is 0.9. Substituting into the formula, the remaining actual current dissipation capacity was calculated to be 9 × 2 × 0.9 = 16.2 kA. The path location and the remaining current dissipation capacity were used as correction parameters, with a correction weight of 1.

[0098] After all correction parameters are extracted, the correction parameters and corresponding correction weights are organized and sent back to the dynamic modeling unit for mountain grounding characteristics. The dynamic modeling unit for mountain grounding characteristics uses a weighted Kalman filter method to iteratively update the mountain current diffusion characteristic map. The parameters of the corresponding grid are smoothly updated according to the correction weights to ensure the stability of the map update. The updated mountain current diffusion characteristic map will be used for the next round of current diffusion weight calculation and surge prediction. This multi-node state control method with time synchronization and confidence judgment can reduce the false alarm rate by more than 70% and the anomaly detection rate to more than 99%. The return of weighted correction parameters can avoid model fluctuations caused by single measurement errors, ensuring that the mountain current diffusion characteristic map is always consistent with the actual field conditions, providing a reliable model basis for subsequent current diffusion optimization and active lightning protection.

[0099] Please see Figure 2 As shown, the second objective of this invention is to provide a construction method for implementing a distributed wind power grounding and lightning protection system in mountainous areas, comprising the following steps: S1. Access topographic data, soil resistivity data, and lightning activity data. After preprocessing, perform fusion calculations to generate a mountain runoff characteristic map. S2. Calculate the current distribution weight allocation instruction for each grounding point based on the mountain current distribution characteristic map, send the current distribution weight allocation instruction to the grounding hardware module, complete the pre-adjustment of the conduction state of the controllable current distribution branch, determine the thunderstorm warning period through multi-source lightning data classification, and send the pre-adjustment result of the current distribution weight to the surge prediction stage during the thunderstorm warning period. S3. Combine the pre-adjustment results of the current dissipation weight with real-time monitoring data to complete the surge prediction calculation. Determine the advance sending time of the instruction based on the inherent conduction action duration of the controllable discharge path, output the trigger instruction to the cable lightning protection module in advance, sort the controllable discharge paths according to the current dissipation weight of the corresponding grounding point, complete the active guidance and discharge of lightning surge, and send the trigger execution status back to the multi-source data fusion modeling stage. S4. Access all node operation data to complete anomaly identification and early warning on the grounding side and lightning protection side. Extract corresponding correction parameters according to different anomaly types, and send the correction parameters back to the multi-source data fusion modeling stage to iteratively update the mountain runoff feature map.

[0100] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely preferred examples and are not intended to limit the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. A distributed wind power grounding and lightning protection system for mountainous areas, electrically connected to distributed wind turbine generators, grounding hardware modules with controllable current dissipation branches, cable lightning protection modules with controllable discharge paths, and field sensing modules, characterized in that... This includes a dynamic modeling unit for mountain grounding characteristics, a collaborative optimization unit for grounding point current dissipation, an active control unit for cable transition lightning protection, and a multi-node state fusion management and control unit. The mountain grounding characteristic dynamic modeling unit receives terrain data, soil resistivity data, and lightning activity data, performs fusion calculations, generates a mountain current dissipation characteristic map, and outputs it synchronously to the grounding point current dissipation collaborative optimization unit and the cable transition lightning protection active control unit. The grounding point current dissipation collaborative optimization unit calculates the current dissipation weight allocation instructions for each grounding point based on the mountain current dissipation characteristic map and sends them to the grounding hardware module for pre-adjusting the conduction status of the controllable current dissipation branch. During the thunderstorm warning period, the pre-adjustment results of the current dissipation weight are simultaneously sent to the cable transition lightning protection active control unit. The active control unit for cable transition lightning protection combines the pre-adjustment results of current dissipation weight with real-time monitoring data to complete surge prediction calculation, outputs trigger commands to the cable lightning protection module in advance to conduct the corresponding discharge path, and sends the trigger execution status back to the dynamic modeling unit for mountain grounding characteristics. The multi-node status fusion control unit accesses the full node operation data to complete anomaly identification and early warning, obtains correction parameters, and sends the correction parameters back to the mountain grounding feature dynamic modeling unit to iteratively update the mountain runoff feature map.

2. The mountain-based distributed wind power grounding and lightning protection system according to claim 1, characterized in that: The topographic data includes basic data on mountain elevation, slope, and aspect obtained from previous surveys, as well as actual topographic change data updated synchronously after each grounding construction, wind turbine location adjustment, and cable route change. Soil resistivity data is collected at fixed intervals by resistivity acquisition devices deployed at different soil depths at each grounding point. During the acquisition process, soil moisture content and soil temperature data at the corresponding points are acquired simultaneously to complete resistivity value correction. Lightning activity data consists of local electric field intensity and lightning proximity signals collected by atmospheric electric field acquisition devices and traveling wave monitoring devices deployed locally at the wind farm, and lightning strike location, lightning current amplitude, and lightning movement path data obtained from the regional lightning location system. All collected topographic data, soil resistivity data, and lightning activity data are preprocessed before being fused and calculated to generate a mountain diffuse current characteristic map.

3. The mountain-based distributed wind power grounding and lightning protection system according to claim 2, characterized in that: By overlaying actual terrain change data onto basic mountain elevation, slope, and aspect data, a three-dimensional mountain terrain grid model covering the entire wind field is generated. Numerically corrected soil resistivity data, combined with soil moisture content and soil temperature at corresponding points, are mapped to the corresponding grid positions in the three-dimensional mountain terrain grid model according to the coordinates of the collected points, generating a soil resistivity distribution layer for the entire region. Local electric field strength, lightning proximity signals, lightning strike location, lightning current amplitude, and lightning movement path data are combined with spatial coordinate and time series matching to calculate the lightning occurrence probability at each grid point, generating a regional lightning risk probability distribution layer. The soil resistivity distribution layer for the entire region and the regional lightning risk probability distribution layer are spatially aligned grid by grid, and each grid point is assigned a corresponding current dissipation capacity weight and lightning protection level. Finally, a dynamic mountain current dissipation characteristic map is generated, which changes with the updated collected data.

4. The mountain-based distributed wind power grounding and lightning protection system according to claim 3, characterized in that: The current dissipation capacity weight and lightning protection level of each grounding point are extracted from the mountain current dissipation feature map. Combined with the inherent rated conduction capacity of the controllable current dissipation branch connected to each grounding point and the current dissipation capacity weight of the corresponding location, the maximum dissipable current dissipation ratio of each grounding point is calculated. The current dissipation priority of each grounding point is determined according to the lightning protection level of the corresponding grid location. The current dissipation weight is evenly distributed according to the maximum dissipable current dissipation ratio of each grounding point. During the thunderstorm warning period, combined with the lightning movement path data, the excess current dissipation weight of grounding points within the lightning movement path coverage area that currently have a current dissipation weight exceeding their own maximum dissipable current dissipation ratio is transferred to the surrounding grounding points that are not within the lightning coverage area and have remaining capacity in the maximum dissipable current dissipation ratio, so as to avoid single-point current dissipation overload. According to the final determined current dissipation weight of each grounding point, the conduction quantity and conduction duration control parameters of the corresponding controllable current dissipation branch are matched and generated. The combination forms the current dissipation weight allocation command and is sent to the grounding hardware module. At the same time, during the thunderstorm warning period, the pre-adjustment result of the current dissipation weight is generated and sent to the cable transition lightning protection active control unit.

5. The mountain-based distributed wind power grounding and lightning protection system according to claim 4, characterized in that: After receiving the current dissipation weight allocation command, the grounding hardware module parses the control parameters for the number of conductions and conduction duration corresponding to each grounding point carried in the command. It then checks the total inherent rated conduction capacity of the currently conducted controllable current dissipation branches at each grounding point against the current dissipation weight assigned to that grounding point, calculating the matching difference between the two. During adjustment, grounding points with a positive matching difference are designated as grounding points where current dissipation capacity needs to be increased. The unused controllable current dissipation branches are sequentially conducted according to their respective current dissipation capacity weights until the total inherent rated conduction capacity of the conducted controllable current dissipation branches matches the assigned current dissipation weight. Grounding points with a negative matching difference are designated as grounding points where current dissipation capacity needs to be reduced. The system sequentially shuts off the already connected controllable current dissipation branches until the total inherent rated conduction capacity matches the allocated current dissipation weight. During thunderstorm warning periods, in addition to the above adjustments, the pre-adjustment process determines the number of reserved redundant conduction branches based on the received current dissipation weight pre-adjustment results. For grounding points receiving transferred current dissipation weights, an additional number of reserved controllable current dissipation branches are connected to reserve current dissipation redundancy capacity. For grounding points transferring current dissipation weights, the system maintains the corresponding number of controllable current dissipation branches connected according to the current dissipation weight requirement. After all controllable current dissipation branches are adjusted, the actual conduction status of each grounding point is verified to be consistent with the current dissipation weight allocation command requirements. A conduction status adjustment completion flag is generated and uploaded to the multi-node status fusion management unit.

6. The mountain-based distributed wind power grounding and lightning protection system according to claim 5, characterized in that: The specific process for determining the thunderstorm warning period is as follows: The system receives real-time data on local electric field strength, lightning proximity signals, lightning movement paths, lightning strike locations, and lightning current amplitude. It then uses this data, combined with a regional lightning risk probability distribution layer, to classify and determine the appropriate level of lightning current. The lightning current amplitude data is composed of direct lightning current monitoring values ​​collected by traveling wave monitoring devices deployed locally at the wind farm and cloud-to-ground lightning current monitoring values ​​transmitted from the regional lightning location system. When the local electric field strength reaches the warning threshold and a lightning proximity signal indicates that a thunderstorm cloud has entered a preset area surrounding the wind farm, a primary thunderstorm warning is generated. A secondary warning is issued when the lightning movement path data shows the thunderstorm's movement direction pointing towards the wind farm area. By combining the time difference of lightning strike locations and the corresponding lightning current amplitude at consecutive moments to calculate the lightning movement speed, and obtaining the expected time for the lightning to reach the wind farm boundary is less than the warning preparation time and the grounding protection range marked by the lightning risk probability distribution layer of the lightning movement path coverage area, it is determined that the thunderstorm warning period has officially begun. Simultaneously, the grounding point current dissipation collaborative optimization unit generates the pre-adjustment result of the current dissipation weight and sends it to the cable transition lightning protection active control unit. When no new lightning strike location is collected within the continuous monitoring time and the local electric field strength falls back below the safety threshold, it is determined that the thunderstorm warning period has ended and the pre-adjustment state is lifted.

7. The mountain-based distributed wind power grounding and lightning protection system according to claim 6, characterized in that: During thunderstorm warning periods, the pre-adjustment results of current dissipation weights are used to summarize the current dissipation weights of all grounding points around the connection point between overhead and buried cables, calculating the total current dissipation capacity around the connection point. Simultaneously, traveling wave signals, real-time local electric field strength, real-time lightning current amplitude, and lightning movement path data collected by traveling wave monitoring devices deployed at the connection point are input. Based on the lightning movement path data and the estimated arrival time of the thunderstorm, the time window for the surge to reach the connection point is preliminarily calculated. The waveform characteristics of the cable traveling wave signal are identified in real time. When the waveform characteristics match the lightning surge waveform, the real-time lightning current amplitude is combined with pre-calibrated electrical current characteristics. The cable transmission attenuation coefficient is used to calculate the actual amplitude of the surge reaching the connection point. The cable transmission attenuation coefficient is a fixed calculation parameter pre-calibrated according to the cable type and laying length. The actual amplitude is compared with the total current dissipation capacity around the connection point. When the actual amplitude exceeds the rated discharge capacity of the normally open discharge path in the cable lightning protection module, the number of additional controllable discharge paths that need to be opened is calculated based on the difference between the actual amplitude and the total current dissipation capacity around the connection point. A trigger command is generated and output to the cable lightning protection module. At the same time, the trigger time, the number of open paths and the actual amplitude are recorded, and the trigger execution status is generated and sent back to the mountain grounding characteristic dynamic modeling unit.

8. The mountain-based distributed wind power grounding and lightning protection system according to claim 7, characterized in that: The specific process of sending the trigger command to the cable surge protection module in advance to activate the corresponding discharge path is as follows: The advance transmission time of the trigger command is calculated based on the inherent conduction duration of the controllable discharge path. The inherent conduction duration of the controllable discharge path is the factory-calibrated fixed time from receiving the trigger command to full conduction. The trigger command is sent at the beginning of the surge arrival time window, ahead of this inherent conduction duration, ensuring that the discharge path is fully conductive when the surge reaches the connection point between the overhead and buried cables. After receiving the trigger command, the cable surge protection module sorts the controllable discharge paths according to their corresponding grounding points from largest to smallest current dissipation weight. Starting from the path corresponding to the first path in the sorted list, it sequentially activates the calculated number of additional controllable discharge paths that need to be activated. During the process, traveling wave monitoring devices deployed at the connection points are used to monitor the potential at the connection points in real time to avoid potential differences when different paths are connected. After all designated paths are connected, the conduction resistance of the connected controllable discharge path is detected. The conduction resistance is the path electrical parameter collected by the traveling wave monitoring device. If there is a fault where the conduction resistance of a path exceeds the set threshold, the backup discharge path reserved in the cable lightning protection module is immediately connected to supplement the discharge capacity. The backup discharge path is a redundant controllable discharge path reserved in the cable lightning protection module. After confirming that all discharge paths are normally connected, the connection completion status is added to the trigger execution status and sent back to the mountain grounding characteristic dynamic modeling unit.

9. The mountain-based distributed wind power grounding and lightning protection system according to claim 1, characterized in that: The multi-node status fusion management and control unit accesses all node operation data to complete anomaly identification and early warning, and the specific process of obtaining correction parameters is as follows: The system connects to the grounding hardware module, which uploads the conduction status adjustment completion flag, the actual conduction status of each controllable current dissipation branch, and the real-time grounding resistance data of each grounding point. It also connects to the cable lightning protection module, which uploads the trigger execution status, the actual conduction status of each controllable discharge path, and the conduction resistance data. Furthermore, it connects to the field sensor module, which uploads the real-time soil resistivity data and lightning activity monitoring data. The real-time grounding resistance data is the grounding resistance value collected by the field sensor module at each grounding point. The system compares the various types of actual operating data with the corresponding command requirements and preset thresholds point by point to identify two types of abnormal states. One type is grounding-side abnormalities, including mismatch between the actual conduction status of the controllable current dissipation branch and the current dissipation weight allocation command requirements, real-time grounding resistance of the grounding point exceeding the threshold, and real-time soil resistivity change exceeding the threshold. The other type is lightning protection-side abnormalities, including mismatch between the actual conduction status of the controllable discharge path and the trigger command requirements, conduction resistance exceeding the threshold, and controllable discharge path not being reset normally after a thunderstorm. Upon detecting an abnormal state, an early warning message containing the location and type of the abnormal point is generated and pushed to the system's associated maintenance alarm terminal. The maintenance alarm terminal is a terminal device associated with the system used to receive abnormal alarm information. Based on the detected abnormal type, corresponding correction parameters are extracted. When the abnormality is a grounding resistance abnormality caused by a change in soil resistivity, the changed soil resistivity value and the corresponding point coordinates are extracted as correction parameters. When the abnormality is a conduction abnormality caused by a fault in a controllable current dissipation branch or controllable discharge path, the location of the faulty branch or path and the remaining actual current dissipation capacity after the fault are extracted as correction parameters. The remaining actual current dissipation capacity is a value calculated based on the inherent rated conduction capacity of the remaining conductive controllable current dissipation branch or controllable discharge path after the fault. When the abnormality is that the deviation between the actual lightning strike location and the predicted location of the lightning movement path exceeds a threshold, the actual lightning strike location and the actual lightning current amplitude are extracted as correction parameters. All correction parameters are then processed and sent back to the mountain grounding feature dynamic modeling unit for iterative updating of the mountain current dissipation feature map.

10. A construction method for implementing a mountain-based distributed wind power grounding and lightning protection system according to any one of claims 1-9, characterized in that: Includes the following steps: S1. Access topographic data, soil resistivity data, and lightning activity data. After preprocessing, perform fusion calculations to generate a mountain runoff characteristic map. S2. Calculate the current distribution weight allocation instruction for each grounding point based on the mountain current distribution characteristic map, send the current distribution weight allocation instruction to the grounding hardware module, complete the pre-adjustment of the conduction state of the controllable current distribution branch, determine the thunderstorm warning period through multi-source lightning data classification, and send the pre-adjustment result of the current distribution weight to the surge prediction stage during the thunderstorm warning period. S3. Combine the pre-adjustment results of the current dissipation weight with real-time monitoring data to complete the surge prediction calculation. Determine the advance sending time of the instruction based on the inherent conduction action duration of the controllable discharge path, output the trigger instruction to the cable lightning protection module in advance, sort the controllable discharge paths according to the current dissipation weight of the corresponding grounding point, complete the active guidance and discharge of lightning surge, and send the trigger execution status back to the multi-source data fusion modeling stage. S4. Access all node operation data to complete the identification and early warning of anomalies on the grounding side and lightning protection side. Extract corresponding correction parameters according to different anomaly types, and send the correction parameters back to the multi-source data fusion modeling stage to iteratively update the mountain runoff feature map.

Images