Microtopographic area rainfall differentiation analysis method and system

Abstract

The invention relates to a micro-topography area rainfall differentiation analysis method and system, and the method specifically comprises the steps: carrying out the three-dimensional topography modeling of a target area, and dividing the target area into a plurality of micro-topography types according to the influence of the target area on the wind speed; the microtopography types comprise windward slopes, leeward slopes, valleys, reservoirs, river valleys and islands; collecting multi-source meteorological data of a target area, wherein the multi-source meteorological data comprises a meteorological station historical rainfall record, satellite remote sensing inversion data, radar echo data and Internet of Things sensor real-time monitoring data; constructing a precipitation correction model based on a machine learning algorithm, and training the precipitation correction model by taking the multi-source meteorological data of the target area as input and the actual precipitation amount as output; wherein the rainfall correction coefficient of each microtopography is calculated according to the type of the microtopography, and the rainfall correction coefficient is dynamically adjusted in combination with a dynamic adjustment mechanism.

Description

Technical Field

[0001] The present invention relates to the technical field of flood prevention for power facilities, and particularly to a method and system for differential analysis of precipitation in micro-topographic regions. Background Art

[0002] Due to terrain undulations, vegetation distribution, and local circulation effects, significant spatial heterogeneity of precipitation often occurs in micro-topographic regions (such as mountains, river valleys, coastal zones, etc.). Traditional precipitation analysis methods are mostly based on large-scale climate models or uniform terrain assumptions, and have the following technical defects: 1. Insufficient characterization of micro-topographic effects: Existing solutions mostly use a single slope or elevation factor to correct precipitation, without considering the combined effects of multiple micro-topographic types (windward slope / leeward slope, valley / ridge, island, etc.); 2. Limited accuracy of data-driven models: Conventional precipitation correction models rely on historical meteorological station data, and have three defects: First, data sparsity: The meteorological station coverage rate in coastal micro-topographic areas is less than 30%, and the distance between rain gauges is generally greater than 10 km; Second, parameter staticity: The weights of meteorological parameters in typhoon seasons and non-typhoon seasons, dry seasons and rainy seasons are not dynamically adjusted, resulting in deviation of correction coefficients; Third, insufficient integration of multi-source data: New data sources such as satellite remote sensing inversion of water vapor flux and real-time wind speed of the Internet of Things are not effectively integrated.

[0003] At the same time, existing precipitation prediction models are mostly based on large-scale climate data, and it is difficult to reflect the local enhancement or weakening effect of micro-topography on rainfall. For example, the rainfall in the mountainous area upstream of a reservoir is usually more than 30% higher than that downstream, but such differences are not considered in traditional designs, resulting in insufficient effectiveness of flood control facilities.

[0004] In a Chinese invention patent document with the publication number of "CN111624682B" in the prior art, a method for quantitative precipitation estimation based on multi-source data fusion is disclosed. It is disclosed that a method based on multi-source big data is used to determine the precipitation falling area, the precipitation retrieved by microwave links is interpolated into discrete grids and accumulated to the △T-minute time scale, the precipitation retrieved by weather radars is interpolated into discrete grids and converted into precipitation on the △T-minute time scale by the time-weighted average method, the precipitation observed by rain gauges on the △T-minute time scale is obtained, interpolation methods are used to establish three types of samples on the △T-minute time scale, corresponding spatio-temporal local models are used for quality control of the three types of samples respectively, and a method of fusing multiple spatio-temporal local models is used to estimate precipitation and accumulate precipitation. The beneficial effects of the present invention are: Through the fusion of multi-source data and multi-models, on the basis of realizing accurate monitoring of the precipitation falling area, the accuracy of quantitative precipitation estimation is further improved.

[0005] In the prior art, the influence of multi-scale data on precipitation prediction is considered, but the influence of various types of micro-topography on rainfall is not considered, and the prediction results are not accurate enough. Summary of the Invention

[0006] To solve the problems existing in the above-mentioned prior art, the present invention proposes a method and system for analyzing precipitation differentiation in micro-topography regions.

[0007] The technical solution of the present invention is as follows:

[0008] On the one hand, the present invention proposes a method for analyzing precipitation differentiation in micro-topography regions, and the specific steps include:

[0009] Based on the three-dimensional terrain modeling of the target area, and dividing the target area into various micro-topography types according to the influence of the target area on the wind speed; the micro-topography types include: windward slope, leeward slope, valley, reservoir, river valley, and island;

[0010] Collect multi-source meteorological data of the target area, including historical precipitation records of meteorological stations, satellite remote sensing inversion data, radar echo data, and real-time monitoring data of Internet of Things sensors;

[0011] Based on the machine learning algorithm, construct a precipitation correction model, and use the multi-source meteorological data of the target area as the input and the actual precipitation as the output to train the precipitation correction model; among them, calculate the precipitation correction coefficients of each micro-topography according to the micro-topography type, and dynamically adjust the precipitation correction coefficients in combination with the dynamic adjustment mechanism;

[0012] Generate a micro-topography precipitation distribution map according to the actual output of the precipitation correction model, and then combine the power grid facility layout in each target area to carry out differential flood control parameter design.

[0013] As a preferred implementation manner, the machine learning method in the step of constructing the precipitation correction model based on the machine learning algorithm is a random forest model, and specific characteristic variables are extracted from the multi-source meteorological data of the target area, including sunshine duration, sea-land breeze conversion frequency, altitude gradient, proportion of surrounding water area, and humidity of the previous day.

[0014] As a preferred implementation manner, the specific calculation of the precipitation correction coefficients of each micro-topography according to the micro-topography type is as follows:

[0015]

[0016] In the formula, H is the relative elevation difference, H0 is the reference elevation, θ is the included angle between the wind direction and the terrain slope, D is the distance from the water source, and α, β, γ, and δ are regression coefficients.

[0017] As a preferred implementation manner, the dynamic adjustment mechanism includes:

[0018] Set up a real-time correction module, access the satellite remote sensing inversion data, and adjust the regression coefficients in real time based on the deviation between the real-time typhoon path and the target typhoon path;

[0019] Set a seasonal weight factor to amplify the calculated precipitation correction coefficient by a preset multiple during the rainy season and reduce the calculated precipitation correction coefficient by a preset multiple during the dry season;

[0020] Optimize multi-source data using spatial interpolation, process sparse meteorological station data based on Kriging interpolation method, and introduce a terrain roughness correction term into the weight function.

[0021] As a preferred embodiment, the differentiated flood control parameters include the substation foundation elevation, drainage ditch capacity, and cable waterproof level.

[0022] On the other hand, the present invention proposes a differential precipitation analysis system for micro-topography areas, including:

[0023] A three-dimensional terrain construction model that models the three-dimensional terrain of the target area and divides the target area into multiple micro-topography types according to the influence of the target area on the wind speed; the micro-topography types include: windward slope, leeward slope, valley, reservoir, river valley, and island;

[0024] A multi-source data acquisition module that acquires multi-source meteorological data of the target area, including historical precipitation records of meteorological stations, satellite remote sensing inversion data, radar echo data, and real-time monitoring data of Internet of Things sensors;

[0025] A precipitation correction dynamic adjustment module that constructs a precipitation correction model based on a machine learning algorithm, trains the precipitation correction model with the multi-source meteorological data of the target area as the input and the actual precipitation as the output; wherein, the precipitation correction coefficients of each micro-topography are calculated according to the micro-topography type, and the precipitation correction coefficients are dynamically adjusted in combination with a dynamic adjustment mechanism;

[0026] A flood control parameter design module that generates a micro-topography precipitation distribution map according to the actual output of the precipitation correction model, and then conducts differentiated flood control parameter design in combination with the power grid facility layout in each target area.

[0027] As a preferred embodiment, the machine learning method in the step of constructing the precipitation correction model based on the machine learning algorithm is a random forest model, and specific characteristic variables are extracted from the multi-source meteorological data of the target area, including sunshine duration, sea-land breeze conversion frequency, altitude gradient, proportion of surrounding water area, and humidity of the previous day.

[0028] As a preferred embodiment, the specific calculation of the precipitation correction coefficients of each micro-topography according to the micro-topography type is as follows:

[0029]

[0030] In the formula, H is the relative elevation difference, H0 is the reference elevation, θ is the angle between the wind direction and the terrain slope, D is the distance from the water source, and α, β, γ, and δ are regression coefficients.

[0031] As a preferred embodiment, the dynamic adjustment mechanism includes:

[0032] Set up a real-time correction module, access satellite remote sensing inversion data, and adjust the regression coefficient in real time based on the deviation between the real-time typhoon path and the typhoon target path;

[0033] Set a seasonal weight factor to amplify the calculated precipitation correction coefficient by a preset multiple during the rainy season and reduce the calculated precipitation correction coefficient by a preset multiple during the dry season;

[0034] Optimize multi-source data using spatial interpolation, process sparse meteorological station data based on Kriging interpolation method, and introduce a terrain roughness correction term into the weight function.

[0035] As a preferred embodiment, the differentiated flood control parameters include the substation foundation elevation, the drainage ditch capacity, and the cable waterproof level.

[0036] The present invention has the following beneficial effects:

[0037] 1. The present invention incorporates various micro-topographic features such as the upslope lifting effect, the downslope sinking air current, the valley constriction effect, and the reservoir water vapor convergence into a unified model, solving the limitation of traditional methods that only consider a single topographic parameter.

[0038] 2. Through the transfer of the Fluent wind speed simulation results, the present invention establishes a quantitative relationship between topographic dynamic parameters (such as the slope angle θ) and precipitation distribution, improving the accuracy compared to pure meteorological models.

[0039] 3. The present invention improves the accuracy of the precipitation correction model through a piecewise correction function, and at the same time dynamically adjusts the precipitation correction coefficient in combination with the dynamic adjustment mechanism, further improving the accuracy of the precipitation correction model.

[0040] 4. The present invention combines the precipitation correction model with the power grid facility layout to design and update traditional flood control parameters, improving the disaster resistance of each region. Description of the Drawings

[0041] Figure 1 It is a schematic flow chart of the steps of the present invention. Detailed Embodiments

[0042] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0043] It should be understood that the step numbers used in the text are only for convenience of description and do not limit the order of execution of the steps.

[0044] It should be understood that the terms used in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the specification of the present invention and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include the plural forms.

[0045] The terms "comprising" and "including" indicate the presence of the described features, wholes, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or their combinations.

[0046] The term "and / or" refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0047] Embodiment 1:

[0048] See Figure 1 , a method for analyzing precipitation differentiation in micro-topography regions, the specific steps including:

[0049] Based on the three-dimensional terrain modeling of the target area, and dividing the target area into multiple micro-topography types according to the influence of the target area on the wind speed; the micro-topography types include: windward slope, leeward slope, valley, reservoir, river valley, island;

[0050] In this embodiment, the investigation area of the embodiment is Fujian, and the influences of several common micro-topographies on the wind speed are as follows:

[0051] 1. Windward slope terrain

[0052] The wind speed on the slope increases: When the main direction of the strong wind forms a large angle or even is almost perpendicular to the ridge line of the mountains in Fujian, and the mountains have a certain height and width and there is no obvious obstruction in the wind direction, the wind speed increases when traveling along the windward slope. For example, in some mountains along the coast of Fujian, when the summer monsoon or typhoon comes, the wind speed on the windward slope side will increase significantly.

[0053] A local strong wind belt is formed: At different positions on the windward slope, the wind speed will also be different. Generally speaking, in the upper and middle parts of the mountain slope, the wind speed is relatively large, and a local strong wind belt may be formed, which has a greater impact on the vegetation, buildings, etc. in this area.

[0054] 2. Perimeter of large water bodies (reservoirs, rivers)

[0055] Wind speed variation on the water surface: There are many large water bodies such as bays and lakes in the coastal areas of Fujian. On the water surface, due to less friction, the wind speed is usually higher than that on land. Moreover, during the day and night, due to the difference in thermal properties between land and water, local sea-land breezes will form, further affecting the wind speed and direction.

[0056] Influence on the wind speed of the surrounding land: On the land side close to the water body, the wind speed will also be affected to a certain extent due to the regulating effect of the water body. For example, when the sea breeze comes ashore, it will increase the wind speed of the coastal land and may penetrate inland for a certain distance.

[0057] 3. Canyon bays

[0058] In some areas of Fujian, such as around Yacheng Bay and Fuxing Bay, there are characteristics of canyon bays. The terrain on both sides of the bay is high. Through the narrow passage effect, air pressure changes will occur, which will increase the wind speed passing through this area and form strong winds.

[0059] The impacts of several common microtopographies on precipitation are as follows:

[0060] 1. Windward slope and leeward slope (the uplifting effect of mountains and the rain shadow effect)

[0061] There are many mountains in Fujian, such as Wuyi Mountains, Jiufeng Mountains, Daiyun Mountains, etc. When warm and moist air currents (such as the air currents brought by the summer monsoon and typhoons) encounter these mountains, the air currents are forced to rise along the mountain slopes. During the rising process, the air expands and cools due to the decrease in air pressure, and the water vapor in it is easy to condense to form rainfall, thus generating abundant orographic rainfall on the windward slope side. For example, when a typhoon lands in Fujian, heavy rain centers often form in the mountainous areas of Xianyou, Yongchun, and Anxi on the east side of Daiyun Mountain and in Nanjing and Pinghe on the east side of Bopingling.

[0062] On the leeward slope side of the mountains, the air currents warm up due to the increase in air pressure during the sinking process, and the water vapor is difficult to condense. Usually, the rainfall is less, forming a rain shadow area. For example, when the air currents cross the Wuyi Mountains, the rainfall in the areas on the west side of the mountains is significantly less than that on the east side.

[0063] 2. River valleys and basins (convergence of water vapor)

[0064] River valleys and basins are relatively low-lying. The water vapor from the surrounding mountains is prone to converge into the basins under the action of gravity. When the water vapor accumulates to a certain extent and there are appropriate uplifting conditions, rainfall may form. However, due to the blocking of the surrounding mountains, the air currents are relatively stable, and the rainfall intensity is generally not as strong as that on the windward slope of the mountains.

[0065] 3. Large reservoirs (increase in water vapor volume, thermal difference, change in wind speed conditions)

[0066] Large reservoirs have a large water surface area, and the evaporation volume of the water body has increased significantly compared to the previous natural state. A large amount of water vapor enters the atmosphere, increasing the air humidity and providing a rich water vapor source for rainfall. When suitable weather systems, such as cold air invasions and low vortices, pass by, the water vapor is easily condensed to form rainfall, increasing the precipitation opportunities in the reservoir area and its surrounding areas.

[0067] The specific heat capacity of the reservoir water body is relatively large, and there are differences in temperature changes compared with the surrounding land. During the day, the water temperature in the reservoir is lower than that of the surrounding land, and the air cools and sinks, forming a local cold high pressure in the center of the reservoir and a local warm low pressure in the surrounding area of the reservoir, accelerating the air flow in the reservoir and its surrounding areas, facilitating the formation of upward movement in the surrounding areas of the reservoir, and then forming clouds and rain; at night, on the contrary, the water temperature in the reservoir is higher than that of the surrounding land, and the air heats up and rises, which also promotes the upward transport of water vapor. This local circulation formed by the thermal difference is conducive to the aggregation and lifting of water vapor, increasing the possibility of rainfall at the reservoir.

[0068] The existence of the reservoir changes the underlying surface conditions. When the air flow passes through the reservoir, it will be affected by the water body, and the speed, direction, etc. of the air flow will change. On the windward shore of the reservoir, the air flow is forced to rise, and the water vapor cools and condenses, easily forming rainfall, just like the lifting effect of mountains on the air flow, increasing the rainfall amount in this area to a certain extent.

[0069] Collect multi-source meteorological data of the target area, including historical precipitation records of meteorological stations, satellite remote sensing inversion data, radar echo data, and real-time monitoring data of Internet of Things sensors;

[0070] In this embodiment, when collecting multi-source data, the water vapor flux in the bay area during typhoons can be retrieved by using Sentinel-1 radar, and 10 ultrasonic rain gauges can be deployed in the upper-middle part of the windward slope (altitude 300 - 400m) to collect raindrop spectrum data in real time and monitor the data in real time.

[0071] Construct a precipitation correction model based on machine learning algorithms, use the multi-source meteorological data of the target area as input, and the actual precipitation as output to train the precipitation correction model; among them, calculate the precipitation correction coefficients of each micro-topography according to the micro-topography type, and dynamically adjust the precipitation correction coefficients in combination with the dynamic adjustment mechanism;

[0072] Generate a micro-topography precipitation distribution map according to the actual output of the precipitation correction model, and then combine the power grid facility layout in each target area to carry out differential flood control parameter design.

[0073] As a preferred embodiment of this embodiment, the machine learning method in the step of constructing the precipitation correction model based on the machine learning algorithm is a random forest model, and specific characteristic variables are extracted from multi-source meteorological data in the target area, including sunshine duration, sea-land breeze conversion frequency, altitude gradient, proportion of surrounding water area, and humidity of the previous day.

[0074] In this embodiment, nine typhoon events from 2016 to 2020 are selected, and a random forest model is established with the measured rainfall as the label.

[0075] As a preferred embodiment of this embodiment, the specific calculation of the precipitation correction coefficient for each micro-topography according to the micro-topography type is as follows:

[0076]

[0077] In the formula, H is the relative elevation difference, H0 is the reference elevation, θ is the included angle between the wind direction and the terrain slope, D is the distance from the water source, and α, β, γ, and δ are regression coefficients.

[0078] As a preferred embodiment of this embodiment, the dynamic adjustment mechanism includes:

[0079] Set up a real-time correction module, access satellite remote sensing inversion data, and adjust the regression coefficients in real time based on the deviation between the real-time typhoon path and the typhoon target path;

[0080] Set up a seasonal weight factor, amplify the calculated precipitation correction coefficient by a preset multiple during the rainy season, and reduce the calculated precipitation correction coefficient by a preset multiple during the dry season;

[0081] Optimize multi-source data using spatial interpolation, process sparse meteorological station data based on Kriging interpolation method, and introduce a terrain roughness correction term into the weight function.

[0082] In this embodiment, the seasonal weight factor dynamically amplifies the correction coefficient in the rainy season by 15% and reduces it by 5% in the dry season.

[0083] As a preferred embodiment of this embodiment, the differentiated flood control parameters include the substation foundation elevation, drainage ditch capacity, and cable waterproof level.

[0084] Embodiment 2:

[0085] A micro-topography area precipitation differentiation analysis system includes:

[0086] A three-dimensional terrain construction model, which models the three-dimensional terrain of the target area and divides the target area into multiple micro-topography types according to the influence of the target area on the wind speed; the micro-topography types include: windward slope, leeward slope, valley, reservoir, river valley, island;

[0087] Multi-source data acquisition module, which acquires multi-source meteorological data of the target area, including historical precipitation records of meteorological stations, satellite remote sensing inversion data, radar echo data, and real-time monitoring data of Internet of Things sensors;

[0088] Precipitation correction dynamic adjustment module, which constructs a precipitation correction model based on machine learning algorithms, uses the multi-source meteorological data of the target area as input, and the actual precipitation as output to train the precipitation correction model; among them, the precipitation correction coefficients of each micro-topography are calculated respectively according to the micro-topography type, and the precipitation correction coefficients are dynamically adjusted in combination with the dynamic adjustment mechanism;

[0089] Flood control parameter design module, which generates a micro-topography precipitation distribution map according to the actual output of the precipitation correction model, and then combines the power grid facility layout in each target area to carry out differential flood control parameter design.

[0090] As a preferred implementation mode of this embodiment, the specific calculation of the precipitation correction coefficients of each micro-topography according to the micro-topography type is as follows:

[0091]

[0092] In the formula, H is the relative elevation difference, H0 is the reference elevation, θ is the included angle between the wind direction and the terrain slope, D is the distance from the water source, and α, β, γ, and δ are regression coefficients.

[0093] As a preferred implementation mode of this embodiment, the dynamic adjustment mechanism includes:

[0094] Set up a real-time correction module, access satellite remote sensing inversion data, and adjust the regression coefficients in real time based on the deviation between the real-time typhoon path and the target typhoon path;

[0095] Set a seasonal weight factor, amplify the calculated precipitation correction coefficient by a preset multiple during the rainy season, and reduce the calculated precipitation correction coefficient by a preset multiple during the dry season;

[0096] Optimize multi-source data using spatial interpolation, process sparse meteorological station data based on Kriging interpolation method, and introduce a terrain roughness correction term into the weight function.

[0097] As a preferred implementation mode of this embodiment, the differential flood control parameters include the substation foundation elevation, the drainage ditch capacity, and the cable waterproof level.

[0098] The above are only embodiments of the present invention, and do not limit the patent scope of the present invention accordingly. Any equivalent structure or equivalent process transformation made by using the content of the specification and drawings of the present invention, or directly or indirectly applied in other related technical fields, shall be equally included in the patent protection scope of the present invention.

Claims

1. A method for analyzing precipitation differentiation in microtopography regions, characterized in that, The specific steps include: Performing three-dimensional terrain modeling of the target area and dividing the target area into multiple micro-topography types according to the influence of the target area on wind speed; the micro-topography types include: windward slope, leeward slope, valley, reservoir, river valley, and island; Collecting multi-source meteorological data of the target area, including historical precipitation records of meteorological stations, satellite remote sensing inversion data, radar echo data, and real-time monitoring data of Internet of Things sensors; Constructing a precipitation correction model based on a machine learning algorithm, using the multi-source meteorological data of the target area as input and the actual precipitation as output to train the precipitation correction model; among them, calculating the precipitation correction coefficients of each micro-topography according to the micro-topography type respectively, and dynamically adjusting the precipitation correction coefficients in combination with a dynamic adjustment mechanism; Generating a micro-topography precipitation distribution map according to the actual output of the precipitation correction model, and then combining the power grid facility layout in each target area to carry out differential flood control parameter design.

2. The method for analyzing precipitation differentiation in microtopography regions according to claim 1, wherein The machine learning method in the step of constructing the precipitation correction model based on the machine learning algorithm is a random forest model, and specific characteristic variables are extracted from the multi-source meteorological data of the target area, including sunshine duration, sea-land breeze conversion frequency, altitude gradient, proportion of surrounding water area, and humidity of the previous day.

3. The method for analyzing precipitation differentiation in microtopography regions according to claim 1, characterized in that The specific calculation of the precipitation correction coefficients of each micro-topography according to the micro-topography type is as follows: In the formula, H is the relative elevation difference, H0 is the reference elevation, θ is the included angle between the wind direction and the terrain slope, D is the distance from the water source, and α, β, γ, and δ are regression coefficients.

4. The method for analyzing precipitation differentiation in micro-topography regions according to claim 3, wherein The dynamic adjustment mechanism includes: Setting up a real-time correction module, accessing satellite remote sensing inversion data, and dynamically adjusting the regression coefficients based on the deviation between the real-time typhoon path and the typhoon target path; Setting a seasonal weight factor to amplify the calculated precipitation correction coefficient by a preset multiple during the rainy season and reduce the calculated precipitation correction coefficient by a preset multiple during the dry season; Optimizing multi-source data using spatial interpolation, processing sparse meteorological station data based on Kriging interpolation method, and introducing a terrain roughness correction term into the weight function.

5. A method for analyzing precipitation differentiation in micro-topography regions according to claim 1, characterized in that, The differential flood control parameters include the foundation elevation of the substation, the capacity of the drainage ditch, and the cable waterproof level.

6. A differential precipitation analysis system for microtopography regions, characterized in that, Including: A three-dimensional terrain construction model, performing three-dimensional terrain modeling of the target area and dividing the target area into multiple micro-topography types according to the influence of the target area on wind speed; The micro-topography types include: windward slope, leeward slope, valley, reservoir, river valley, and island; A multi-source data collection module, collecting multi-source meteorological data of the target area, including historical precipitation records of meteorological stations, satellite remote sensing inversion data, radar echo data, and real-time monitoring data of Internet of Things sensors; A precipitation correction dynamic adjustment module, constructing a precipitation correction model based on a machine learning algorithm, using the multi-source meteorological data of the target area as input and the actual precipitation as output to train the precipitation correction model; among them, calculating the precipitation correction coefficients of each micro-topography according to the micro-topography type respectively, and dynamically adjusting the precipitation correction coefficients in combination with a dynamic adjustment mechanism; A flood control parameter design module, generating a micro-topography precipitation distribution map according to the actual output of the precipitation correction model, and then combining the power grid facility layout in each target area to carry out differential flood control parameter design.

7. The precipitation differential analysis system for micro-topography regions according to claim 6, characterized in that, The machine learning method in the steps of constructing the precipitation correction model based on the machine learning algorithm is the random forest model, and specific characteristic variables are extracted from multi-source meteorological data in the target area, including sunshine duration, sea-land breeze conversion frequency, altitude gradient, proportion of surrounding water area, and humidity of the previous day.

8. The precipitation differential analysis system for micro-topography regions according to claim 6, characterized in that, Specifically, the precipitation correction coefficients of each micro-topography are calculated according to the micro-topography type as follows: In the formula, H is the relative elevation difference, H0 is the reference elevation, θ is the angle between the wind direction and the terrain slope, D is the distance from the water source, and α, β, γ, and δ are regression coefficients.

9. A micro-topography area precipitation differentiation analysis system according to claim 8, characterized in that, The dynamic adjustment mechanism includes: Setting up a real-time correction module to access satellite remote sensing inversion data and adjust the regression coefficients in real time based on the deviation between the real-time typhoon path and the typhoon target path; Setting a seasonal weight factor to magnify the calculated precipitation correction coefficient by a preset multiple in the rainy season and reduce the calculated precipitation correction coefficient by a preset multiple in the dry season; Optimizing multi-source data by spatial interpolation, processing sparse meteorological station data based on Kriging interpolation method, and introducing a terrain roughness correction term into the weight function.

10. A micro-topography area precipitation differentiation analysis system according to claim 6, characterized in that, The differentiated flood control parameters include the substation foundation elevation, drainage ditch capacity, and cable waterproof grade.

Images