Short-time early warning system for urban low-lying areas prone to waterlogging based on DSM and rainfall real-time monitoring

Abstract

A short-time early warning system for urban low-lying areas prone to waterlogging based on DSM and rainfall real-time monitoring comprises the steps that 1, an initial flight path is set for an unmanned aerial vehicle carrying imaging equipment, image data collection is conducted on a monitoring area, data noisy points are screened, the initial flight path is optimized after the data noisy points are screened, and an optimized flight path is obtained; step 2, executing optimization of a flight path by carrying a laser radar unmanned aerial vehicle, establishing a DSM model through laser radar acquisition data, determining waterlogging-prone areas, and calculating a waterlogging area volume Vmax and a waterlogging area catchment area A of each waterlogging-prone area; 3, the waterlogging-prone area is matched with the collected image data, the land surface coverage type of the waterlogging-prone area is determined through a deep learning matching algorithm mode, and the infiltration rate fi is determined through the land surface coverage type of the waterlogging-prone area, and the method has the advantages that the DSM model is rapidly established, the model accuracy is improved by 80% or above, the modeling efficiency is improved by 50% or above, and the method is suitable for large-scale popularization and application. DSM updating within 1-3 days of sudden topographic change is realized.

Description

Technical Field

[0001] This invention relates to a short-term early warning system for urban flood-prone low-lying areas based on DSM and real-time rainfall monitoring. Background Technology

[0002] Low-lying urban areas prone to flooding (such as underpasses, low-lying sections of old city streets, and entrances to underground spaces) are high-risk areas for urban flooding during the flood season. The rapid and insidious formation of water in these areas can easily lead to traffic disruptions, vehicle damage, and even casualties, posing a serious threat to urban operational safety. With the intensification of global climate change and the increasing frequency of extreme weather events such as short-duration heavy rainfall and typhoons, traditional early warning methods for flood-prone areas are no longer sufficient to meet the short-term warning requirements of "early detection, early warning, and early response." There is an urgent need to overcome existing technological bottlenecks and improve the accuracy and timeliness of early warning systems.

[0003] Existing early warning systems rely heavily on traditional digital elevation models (DEMs) or manual surveying data to depict the terrain of flood-prone, low-lying areas. DEM data only reflects the elevation of the exposed ground and does not include information on surface attachments such as buildings, trees, and road railings. This makes it difficult to accurately identify key risk areas such as "micro-low-lying alleys between buildings" and "slope change points at underpass entrances." During heavy rains, the depth of water accumulation in these potholes can far exceed the warning expectations, leading to secondary accidents.

[0004] Existing short-term warning systems mostly use "empirical threshold methods" (such as setting "an hourly rainfall of ≥50mm will trigger an early warning") or simple linear regression models. They have not established a dynamic coupling relationship between rainfall intensity, rainfall duration and water depth and water accumulation speed in low-lying areas, resulting in low warning accuracy and high false alarm and missed alarm rates. Summary of the Invention

[0005] The purpose of this invention is to provide a short-term early warning system for urban flood-prone low-lying areas based on DSM and real-time rainfall monitoring, which can achieve rapid and accurate response to flood risks in low-lying areas in a short time.

[0006] A short-term early warning system for urban flood-prone low-lying areas based on DSM and real-time rainfall monitoring, including Step 1: Set an initial flight path for the drone equipped with imaging equipment, collect image data of the monitoring area, and filter out data noise. After filtering out data noise, optimize the initial flight path to obtain an optimized flight path. Step two involves using a drone equipped with lidar to optimize flight paths, collecting data via lidar to build a DSM model, identifying flood-prone areas, and calculating the flood volume V for each flood-prone area. max The catchment area of ​​the flood-prone area is A; Step 3: Match flood-prone areas with the collected image data. Use a deep learning matching algorithm to determine the land cover type of the flood-prone areas, and then determine the infiltration rate f based on the land cover type. i ; Step 4: Calculate the flow capacity Q using municipal pipeline network information. The calculation formula is Q=A. t ×v, where A t Where v is the cross-sectional area of ​​the pipe, and v is the design flow velocity. Step 5, calculate the volume V of the flood-prone area. max 1. Catchment area A of the flood-prone area; 2. Infiltration rate f i Substituting the flow capacity Q into the formula, the critical rainfall P for flooding is calculated. c The formula is:

[0007] Where t is the duration of rainfall, which can be set to a fixed value. In short-term warning applications, t is set to 0.5h. Step 6: Calculate the critical rainfall threshold P for flooding. c Compared with the actual rainfall P r Compare, when P c Less than or equal to P r Then a flood warning will be issued.

[0008] Furthermore, in step two, after identifying the flood-prone area, its geometric center is determined. A first warning circle with a radius of 10 km, a second warning circle with a radius of 20 km, and a third warning circle with a radius of 30 km are set up with the geometric center as the center. Each warning circle is equipped with at least four real-time rainfall monitoring devices.

[0009] Furthermore, the real-time rainfall monitoring devices are arranged at 360-degree intervals.

[0010] The advantages of this invention are: 1. By using drones equipped with imaging devices, non-target areas can be quickly and accurately screened, making it easier for drones equipped with LiDAR to build DSM models more accurately and quickly. Compared with the method of directly using LiDAR drones for data collection, the model accuracy is improved by more than 80%, the modeling efficiency is improved by more than 50%, and the DSM can be updated within 1-3 days for sudden terrain changes.

[0011] 2. The calculation of critical rainfall for flooding is more accurate, fully taking into account the volume V of the flood-prone area. max 1. Catchment area A of the flood-prone area; 2. Infiltration rate f iThe flow capacity Q enables more accurate calculation of critical rainfall levels for flooding. "Surface cover classification data" is overlaid on the DSM (Digital Subsurface Survey). The area is divided into four categories: "paved roads, green spaces, permeable pavements, and water bodies," with different "infiltration coefficients" assigned to each category. A water accumulation prediction model is embedded in each category, allowing the model to calculate water depth based on the infiltration rate of different cover types, reducing the prediction error to within ±5cm.

[0012] 3. By setting up graded real-time rainfall monitoring devices around flood-prone areas, the probability of flooding can be accurately determined in a short time, without any missed or incorrect judgments. This allows for precise identification of flood-prone areas and facilitates the implementation of emergency measures. Attached Figure Description

[0013] Figure 1 This is a schematic diagram showing the distribution of flood-prone areas and real-time rainfall monitoring devices.

[0014] Figure 2 This is a schematic diagram for calculating the geometric center of a flood-prone area.

[0015] Figure 3 This is a flowchart illustrating a short-term early warning system for urban flood-prone low-lying areas based on DSM and real-time rainfall monitoring.

[0016] Figure 4 This is a schematic diagram of the calculation process for the flood threshold calculation model.

[0017] Attached reference numerals: 1. Real-time rainfall monitoring device; 2. Third warning zone; 3. Flood-prone area; 4. Second warning zone; 5. First warning zone; 6. Geometric center. Detailed Implementation

[0018] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0019] The short-term early warning system for urban flood-prone low-lying areas based on DSM and real-time rainfall monitoring includes the following steps: First, by setting an initial flight path for a drone equipped with imaging equipment, image data is collected from the monitoring area and data noise is filtered out. After filtering out the data noise, the initial flight path is optimized to obtain an optimized flight path. Step two involves using a drone equipped with lidar to optimize flight paths, collecting data via lidar to build a DSM model, identifying flood-prone areas, and calculating the flood volume V for each flood-prone area. max The catchment area of ​​the flood-prone area is A; Step 3: Match flood-prone areas with the collected image data. Use a deep learning matching algorithm to determine the land cover type of the flood-prone areas, and then determine the infiltration rate f based on the land cover type. i ; Step 4: Calculate the flow capacity Q using municipal pipeline network information. The calculation formula is Q=A. t ×v, where A t Where v is the cross-sectional area of ​​the pipe, and v is the design flow velocity. Step 5, calculate the volume V of the flood-prone area. max 1. Catchment area A of the flood-prone area; 2. Infiltration rate f i Substituting the flow capacity Q into the formula, the critical rainfall P for flooding is calculated. c The formula is:

[0020] Where t is the duration of rainfall, which can be set to a fixed value. In short-term warning applications, t is set to 0.5h. Step 6: Calculate the critical rainfall threshold P for flooding. c Compared with the actual rainfall P r Compare, when P c Less than or equal to P r A flood warning will then be issued. In step two, after identifying the flood-prone area, its geometric center is determined. Using this geometric center as the center, three warning circles are established: a first warning circle with a radius of 10 km, a second warning circle with a radius of 20 km, and a third warning circle with a radius of 30 km. Each warning circle is equipped with at least four real-time rainfall monitoring devices. These monitoring devices are arranged at 360-degree equidistant intervals.

[0021] Catchment area of ​​flood-prone areas: Using the "Digital Elevation Watershed Analysis (D8 Algorithm)" of DSM, the direction of water flow and the catchment path are extracted, and the area of ​​all areas that eventually flow into flood-prone areas is delineated. The area of ​​this range is then calculated using GIS software.

[0022] Maximum volume of flood-prone areas: Use DSM data to determine the "minimum elevation" (the starting point of water accumulation) and "warning elevation" (such as road surface elevation + 0.15m, i.e., water depth of 15cm, affecting pedestrian passage).

[0023] Infiltration rate: The infiltration rate varies depending on the type of land cover, which needs to be fully considered to avoid excessive calculation errors. For example, the infiltration rate of hardened pavement is about 0.008, while that of green space is about 0.04, etc.

[0024] Flow capacity: Calculated based on design values. If pipe network aging and blockage are further considered, a more accurate estimate can be made.

[0025] Real-time rainfall monitoring devices include rain gauges. Rain gauges within the warning range periodically transmit real-time rainfall data to the back-end control center, and calculations are performed based on the maximum rainfall value. Setting t to 0.5 hours allows for monitoring of sudden rainstorms and flooding within 30 minutes.

[0026] The first alert zone has a radius of 10 km, the second alert zone has a radius of 20 km, and the third alert zone has a radius of 30 km, which correspond to the first, second and third alert levels, and different response methods are adopted for different alert levels.

[0027] The above description discloses only preferred embodiments of the present invention and should not be construed as limiting the scope of the present invention. Therefore, equivalent variations made in accordance with the claims of the present invention are still within the scope of the present invention.

Claims

1. A short-term early warning system for urban flood-prone low-lying areas based on DSM and real-time rainfall monitoring, characterized in that: include Step 1: Set an initial flight path for the drone equipped with imaging equipment, collect image data of the monitoring area, and filter out data noise. After filtering out data noise, optimize the initial flight path to obtain an optimized flight path. Step two involves using a drone equipped with lidar to optimize flight paths, collecting data via lidar to build a DSM model, identifying flood-prone areas, and calculating the flood volume V for each flood-prone area. max The catchment area of ​​the flood-prone area is A; Step 3: Match flood-prone areas with the collected image data. Use a deep learning matching algorithm to determine the land cover type of the flood-prone areas, and then determine the infiltration rate f based on the land cover type. i ; Step 4: Calculate the flow capacity Q using municipal pipeline network information. The calculation formula is Q=A. t ×v, where A t Where v is the cross-sectional area of ​​the pipe, and v is the design flow velocity. Step 5, calculate the volume V of the flood-prone area. max 1. Catchment area A of the flood-prone area; 2. Infiltration rate f i Substituting the flow capacity Q into the formula, the critical rainfall P for flooding is calculated. c The formula is: ； Where t is the duration of rainfall, which can be set to a fixed value. In short-term warning applications, t is set to 0.5h. Step 6: Calculate the critical rainfall threshold P for flooding. c Compared with the actual rainfall P r Compare, when P c Less than or equal to P r Then a flood warning will be issued.

2. The short-term early warning system for urban flood-prone low-lying areas based on DSM and real-time rainfall monitoring as described in claim 1, characterized in that: In step two, after identifying the flood-prone area, determine its geometric center. Using the geometric center as the center, set up a first warning circle with a radius of 10 km, a second warning circle with a radius of 20 km, and a third warning circle with a radius of 30 km. Each warning circle is equipped with at least four real-time rainfall monitoring devices.

3. The short-term early warning system for urban flood-prone low-lying areas based on DSM and real-time rainfall monitoring as described in claim 2, characterized in that: The real-time rainfall monitoring devices are arranged at equal intervals in a 360-degree pattern.

Images