A method for predicting urban surface water accumulation points based on large model technology

Abstract

This invention provides a method for predicting urban surface water accumulation points based on large-scale model technology, comprising: collecting facility data corresponding to each drainage facility in the city to be tested; establishing an urban drainage model of the city to be tested by combining the city's topographic data; training the urban drainage model with precipitation to obtain urban drainage volume information and urban water accumulation information corresponding to different precipitation intensities; analyzing the regional water accumulation corresponding to each urban area of ​​the city to be tested based on the urban drainage volume information and the urban water accumulation information; determining the water accumulation distribution information of the city to be tested; identifying several water accumulation points corresponding to different precipitation intensities in the city to be tested; and generating a precipitation-water accumulation point distribution map of the city to be tested. This invention can accurately predict the entire process of urban surface water accumulation, helping emergency personnel to take preventive measures in advance, thereby effectively reducing the losses caused by urban rainfall and water accumulation.

Description

Technical Field

[0001] This invention relates to the field of urban drainage technology, and in particular to a method for predicting urban surface water accumulation points based on large model technology. Background Technology

[0002] With urban development, the impermeable surface area of ​​urban areas has increased, making flooding a common occurrence due to rainfall. During periods of heavy rainfall, excessive pressure on drainage networks prevents timely drainage, leading to water accumulation in low-lying areas and severely impacting urban traffic and residents' lives. Furthermore, pipelines that have been in operation for many years are in disrepair. Therefore, it is essential and urgent to understand the current operational status of urban stormwater drainage networks using numerical simulation to optimize and expand them appropriately. Currently, mathematical models are being used to simulate the response of urban surface runoff to rainfall events for urban flood control planning and network optimization. However, existing models generally suffer from problems such as high computational load, long computation time, and high computational power requirements.

[0003] Therefore, this invention provides a method for predicting urban surface water accumulation points based on large model technology. Summary of the Invention

[0004] This invention provides a method for predicting urban surface water accumulation points based on large model technology. It can accurately predict the entire process of urban surface water accumulation, helping emergency personnel to take preventive measures in advance, thereby effectively reducing the losses caused by urban rainfall and water accumulation.

[0005] This invention provides a method for predicting urban surface water accumulation points based on large model technology, including:

[0006] Step 1: Collect facility data for each drainage facility in the city to be tested, and combine it with the topographic data of the city to be tested to establish an urban drainage model for the city to be tested;

[0007] Step 2: Train the urban drainage model using each preset precipitation intensity data to obtain the urban drainage volume and urban water accumulation information of the city under test under different precipitation intensity data;

[0008] Step 3: Analyze the regional water accumulation of each urban area in the city to be tested based on the urban drainage information and the urban water accumulation information, and determine the water accumulation distribution information of the city to be tested based on the regional water accumulation.

[0009] Step 4: Based on the water accumulation distribution information, determine several water accumulation points corresponding to different precipitation intensities in the city to be tested, and generate a precipitation-water accumulation point distribution map of the city to be tested.

[0010] In one feasible approach

[0011] When precipitation occurs in the city to be tested, a real-time precipitation distribution map of the city to be tested is established based on the real-time precipitation intensity corresponding to each urban area.

[0012] Find the high-frequency water accumulation points for each urban area under the corresponding real-time precipitation intensity in the precipitation-water accumulation distribution map.

[0013] Using the real-time precipitation distribution map, real-time water accumulation monitoring is performed on each of the high-frequency water accumulation points to obtain the first real-time water accumulation volume corresponding to each of the high-frequency water accumulation points in the city under test.

[0014] Water accumulation points are sampled for each urban area, and the real-time water accumulation is monitored for each sampled water accumulation point using the real-time precipitation distribution map to obtain the second real-time water accumulation volume corresponding to each sampled water accumulation point in the city to be tested.

[0015] The real-time water accumulation information of the city to be tested is determined and displayed based on the first real-time water accumulation volume corresponding to each of the first high-frequency water accumulation points and the second real-time water accumulation volume corresponding to each of the sampling water accumulation points.

[0016] In one feasible approach

[0017] Also includes:

[0018] Based on the real-time water accumulation information, find several actual water accumulation points in the city to be tested whose real-time water accumulation is higher than the standard water accumulation;

[0019] Establish a water accumulation distribution matrix for the city to be tested based on the actual distribution of water accumulation points.

[0020] The difference between the real-time water accumulation and the standard water accumulation for each actual water accumulation point is input into the water accumulation distribution matrix to obtain the water accumulation information matrix of the city to be tested.

[0021] Singular value decomposition is performed on the water accumulation information matrix to obtain the dimensionality-reduced data corresponding to each water accumulation difference;

[0022] Based on the data trend of the dimensionality-reduced data captured by the preset linear regression model, the water accumulation trend corresponding to each actual water accumulation point is obtained.

[0023] Based on the water accumulation trend, assign a corresponding number of security personnel to each actual water accumulation point.

[0024] In one feasible approach

[0025] Step 1 includes:

[0026] Step 11: Statistically analyze the drainage facilities included in the city to be tested, obtain the facility specification data corresponding to each drainage facility, and obtain the setting location data of each drainage facility in the city to be tested. Construct facility data corresponding to each drainage facility based on the facility specification data and the setting location data.

[0027] Step 12: Obtain the topographic data of the city to be tested, establish the surface undulation information of the city to be tested based on the topographic data, determine several water flow directions of the city to be tested based on the undulation information, and establish the drainage range and drainage direction corresponding to each drainage facility based on the facility data;

[0028] Step 13: Construct a virtual drainage sub-model for each drainage facility based on the drainage range and drainage direction of the same drainage facility, establish a virtual direction sub-model for the city to be tested based on each water flow direction, analyze the water basin length and water self-consumption corresponding to each virtual drainage direction in the virtual direction sub-model using the virtual drainage sub-model, and establish a virtual domain length sub-model for the city to be tested based on the water basin length.

[0029] Step 14: Based on the topographic data, establish a three-dimensional city image of the city to be tested, map the virtual drainage sub-model, the virtual direction sub-model, and the virtual domain length sub-model onto the three-dimensional city image respectively, and adjust the mapping results using the water self-consumption rate to generate the city drainage model of the city to be tested.

[0030] In one feasible approach

[0031] Step 2 includes:

[0032] Step 21: Retrieve historical precipitation information of the geographical location of the city to be tested, determine several first preset precipitation intensity data based on the historical precipitation information, search for several rainstorm disaster precipitation information in the big data, determine several second preset precipitation intensity data based on the rainstorm disaster precipitation information, and establish a precipitation training database for the city to be tested based on the first precipitation intensity data and the second preset precipitation intensity data.

[0033] Step 22: Use each training data in the precipitation training database to perform time-based iterative precipitation training on the urban drainage model to obtain several training results corresponding to each training data. Based on the training results, establish the time-domain drainage information and time-domain water accumulation information of the urban drainage model under each training data.

[0034] Step 23: Determine the training precipitation intensity corresponding to each iteration cycle based on the number of iteration cycles corresponding to each training data, construct a precipitation information axis, and use the precipitation information axis corresponding to the same training data to dynamically analyze the time-domain drainage information and time-domain water accumulation information to obtain the first drainage volume and first water accumulation volume corresponding to the city under test under different time-domain precipitation intensities.

[0035] Step 24: Based on the precipitation training database, establish several dynamic precipitation intensity data. Use the dynamic precipitation intensity data to train the urban drainage model to obtain the second drainage volume and second water accumulation volume corresponding to the urban drainage model under different dynamic precipitation intensities. Based on several first drainage volumes and second drainage volumes, establish the urban drainage information of the city to be tested. Based on several first water accumulation volumes and second water accumulation volumes, establish the urban water accumulation information of the city to be tested.

[0036] In one feasible approach

[0037] Also includes:

[0038] Based on the urban waterlogging information, several information points in the city to be tested where the urban waterlogging volume is greater than the standard waterlogging volume are determined;

[0039] Based on the urban waterlogging information, a training precipitation pattern corresponding to each information point is determined, and a warning trigger label for the information point is established based on the training precipitation pattern.

[0040] When the current precipitation pattern of the city to be tested is the same as the training precipitation pattern, the warning trigger tag is activated, a warning message is generated and displayed.

[0041] In one feasible approach

[0042] Step 3 includes:

[0043] Step 31: Establish several drainage flow paths for the city to be tested based on the urban drainage information corresponding to each rainfall intensity data, and establish several water accumulation flow paths for the city to be tested under the corresponding rainfall intensity data based on the urban water accumulation information.

[0044] Step 32: Divide the city to be tested into several urban areas according to the urban planning of the city to be tested, and obtain the first overlap information between each urban area and different drainage flow paths, and obtain the second overlap information between each urban area and different water accumulation flow paths.

[0045] Step 33: Based on the first overlap information and the second overlap information, determine the corresponding regional water accumulation of the urban area under the corresponding precipitation intensity data, and based on the first overlap information and the second overlap information, determine the water flow deposition characteristics of the corresponding urban area;

[0046] Step 34: Cities with water accumulation exceeding the preset standard water accumulation are designated as waterlogged cities. The water accumulation characteristics and the water accumulation volume are used to determine and display the water accumulation distribution information of each waterlogged city under the corresponding precipitation intensity data.

[0047] In one feasible approach

[0048] Step 4 includes:

[0049] Step 41: Using the water distribution information, establish several water accumulation points in the city under test corresponding to different precipitation intensities, and statistically analyze the water flow information corresponding to each water accumulation point;

[0050] Step 42: Draw the dynamic water flow corresponding to each urban area based on the water accumulation points corresponding to each rainfall intensity and the water flow information;

[0051] Step 43: Determine the location of the water flow collection in each corresponding urban area based on the dynamic water flow, and generate a precipitation-water accumulation point distribution map of the city to be tested.

[0052] In one feasible approach

[0053] Also includes:

[0054] The city drainage model of the city to be tested is updated based on the real-time facility construction information of the city to be tested.

[0055] In one feasible approach

[0056] Also includes:

[0057] Based on the precipitation-water accumulation point distribution map, establish and display the corresponding protection deployment plan for each precipitation intensity.

[0058] The beneficial effects of the above technical solution are as follows: To achieve early prevention and avoid unnecessary trouble caused by water accumulation in the city, firstly, an urban drainage model of the city to be tested is established based on the facility data and topographic data of the city to be tested. Then, the urban drainage model is trained using precipitation data of different intensities to determine the urban drainage volume and urban water accumulation under the influence of precipitation of different intensities. This allows for the analysis of the regional water accumulation corresponding to each urban area in the city to be tested and the determination of the water accumulation distribution information of the city to be tested. Furthermore, it allows for the analysis of the water accumulation points corresponding to different precipitation intensities in the city to be tested, thereby generating a precipitation-water accumulation point distribution map of the city. In this way, when it rains in the city to be tested, the water accumulation points in the city can be located according to the precipitation intensity, facilitating timely handling of water accumulation points by relevant personnel and ensuring the normal operation of the city.

[0059] Other features and advantages of the invention will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the written description and the accompanying drawings.

[0060] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0061] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0062] Figure 1 This is a schematic diagram illustrating the workflow of a method for predicting urban surface water accumulation points based on large model technology in an embodiment of the present invention.

[0063] Figure 2 This is a schematic diagram of step 1 of a method for predicting urban surface water accumulation points based on large model technology in an embodiment of the present invention. Detailed Implementation

[0064] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0065] Example 1

[0066] This embodiment provides a method for predicting urban surface water accumulation points based on large model technology, such as... Figure 1 As shown, it includes:

[0067] Step 1: Collect facility data for each drainage facility in the city to be tested, and combine it with the topographic data of the city to be tested to establish an urban drainage model for the city to be tested;

[0068] Step 2: Train the urban drainage model using each preset precipitation intensity data to obtain the urban drainage volume and urban water accumulation information of the city under test under different precipitation intensity data;

[0069] Step 3: Analyze the regional water accumulation of each urban area in the city to be tested based on the urban drainage information and the urban water accumulation information, and determine the water accumulation distribution information of the city to be tested based on the regional water accumulation.

[0070] Step 4: Based on the water accumulation distribution information, determine several water accumulation points corresponding to different precipitation intensities in the city to be tested, and generate a precipitation-water accumulation point distribution map of the city to be tested.

[0071] In this example, drainage facilities include: manholes, drainage channels, and rainwater collection stations, etc.

[0072] In this example, the topographic data represents the elevation and undulation data of the city to be measured;

[0073] In this example, the preset precipitation intensity data refers to data that is set in advance to represent precipitation of different intensities;

[0074] In this example, precipitation training represents the process of simulating the impact of precipitation of different intensities on the city under test;

[0075] In this example, the water distribution information represents the distribution of water in the city being tested;

[0076] In this example, the precipitation-water accumulation point distribution represents the distribution of water accumulation points in the city under different intensities of precipitation.

[0077] The working principle and beneficial effects of the above technical solution are as follows: To achieve early prevention and avoid unnecessary trouble caused by water accumulation in the city, a city drainage model is first established based on the facility and topographic data of the city to be tested. Then, the city drainage model is trained using precipitation data of different intensities to determine the city's drainage volume and water accumulation under the influence of precipitation of different intensities. This allows for the analysis of the regional water accumulation corresponding to each urban area in the city to be tested and the determination of the water accumulation distribution information of the city. Furthermore, it analyzes the water accumulation points corresponding to different precipitation intensities in the city to be tested, thereby generating a precipitation-water accumulation point distribution map of the city. In this way, when it rains in the city to be tested, the water accumulation points can be located according to the precipitation intensity, facilitating timely handling of water accumulation points by relevant personnel and ensuring the normal operation of the city.

[0078] Example 2

[0079] Based on Example 1, the method for predicting urban surface water accumulation points based on large model technology further includes:

[0080] When precipitation occurs in the city to be tested, a real-time precipitation distribution map of the city to be tested is established based on the real-time precipitation intensity corresponding to each urban area.

[0081] Find the high-frequency water accumulation points for each urban area under the corresponding real-time precipitation intensity in the precipitation-water accumulation distribution map.

[0082] Using the real-time precipitation distribution map, real-time water accumulation monitoring is performed on each of the high-frequency water accumulation points to obtain the first real-time water accumulation volume corresponding to each of the high-frequency water accumulation points in the city under test.

[0083] Water accumulation points are sampled for each urban area, and the real-time water accumulation is monitored for each sampled water accumulation point using the real-time precipitation distribution map to obtain the second real-time water accumulation volume corresponding to each sampled water accumulation point in the city to be tested.

[0084] The real-time water accumulation information of the city to be tested is determined and displayed based on the first real-time water accumulation volume corresponding to each of the first high-frequency water accumulation points and the second real-time water accumulation volume corresponding to each of the sampling water accumulation points.

[0085] In this example, high-frequency water accumulation points refer to the water accumulation points in the city under real-time precipitation intensity, which are determined by training the urban drainage model. In other words, the water accumulation points can be found in the precipitation-water accumulation distribution map.

[0086] In this example, the first real-time water accumulation volume represents the water accumulation volume corresponding to the high-frequency water accumulation point, and the second real-time water accumulation volume represents the water accumulation volume corresponding to the sampling water accumulation point;

[0087] In this example, the number of water accumulation points sampled is twice the number of high-frequency water accumulation points, and the water accumulation points sampled do not overlap with the high-frequency water accumulation points.

[0088] The working principle and beneficial effects of the above technical solution are as follows: When precipitation occurs in the city to be tested, several high-frequency water accumulation points in the city are determined based on the precipitation-water accumulation distribution map. Then, real-time water accumulation monitoring is carried out on the high-frequency water accumulation points. In order to avoid randomness and improve the reliability of monitoring, water accumulation points are sampled in the city to be tested, and then the sampled water accumulation points are also monitored in real time. In this way, real-time water accumulation information of the city to be tested can be established based on the real-time water accumulation monitoring results. Not only are the high-frequency water accumulation points analyzed, but also the overall water accumulation situation of the city is analyzed.

[0089] Example 3

[0090] Based on Example 2, the method for predicting urban surface water accumulation points based on large model technology further includes:

[0091] Based on the real-time water accumulation information, find several actual water accumulation points in the city to be tested whose real-time water accumulation is higher than the standard water accumulation;

[0092] Establish a water accumulation distribution matrix for the city to be tested based on the actual distribution of water accumulation points.

[0093] The difference between the real-time water volume and the standard water volume corresponding to each actual water accumulation point is input into the water accumulation distribution matrix to obtain the water accumulation information matrix of the city to be tested.

[0094] Singular value decomposition is performed on the water accumulation information matrix to obtain the dimensionality-reduced data corresponding to each water accumulation difference;

[0095] Based on the data trend of the dimensionality-reduced data captured by the preset linear regression model, the water accumulation trend corresponding to each actual water accumulation point is obtained.

[0096] Based on the water accumulation trend, assign a corresponding number of security personnel to each actual water accumulation point.

[0097] In this example, the standard water accumulation is less than or equal to 15mm;

[0098] In this example, the water accumulation distribution matrix is ​​a matrix used to represent the location of water accumulation points, established based on the actual location distribution relationship of the water accumulation points.

[0099] In this example, the water accumulation trend represents the trend of change in the amount of water accumulated at an actual water accumulation point;

[0100] In this example, the dimensionality-reduced data represents the data obtained after mapping the water accumulation difference to a lower-dimensional space;

[0101] In this example, singular value decomposition represents the process of decomposing the water accumulation information matrix into the product of three matrices;

[0102] In this example, the predefined linear regression model represents a mathematical model used to analyze the linear relationship between dimensionality-reduced data.

[0103] The working principle and beneficial effects of the above technical solution are as follows: by using real-time water accumulation information to determine several actual water accumulation points in the city to be tested, and then using a matrix to analyze the water accumulation trend corresponding to each actual water accumulation point in the city to be tested, thereby assigning corresponding support personnel to the water accumulation points and implementing effective dredging and support.

[0104] Example 4

[0105] Based on Example 1, the method for predicting urban surface water accumulation points based on large model technology, such as... Figure 2 Step 1 as shown includes:

[0106] Step 11: Statistically analyze the drainage facilities included in the city to be tested, obtain the facility specification data corresponding to each drainage facility, and obtain the setting location data of each drainage facility in the city to be tested. Construct facility data corresponding to each drainage facility based on the facility specification data and the setting location data.

[0107] Step 12: Obtain the topographic data of the city to be tested, establish the surface undulation information of the city to be tested based on the topographic data, determine several water flow directions of the city to be tested based on the undulation information, and establish the drainage range and drainage direction corresponding to each drainage facility based on the facility data;

[0108] Step 13: Construct a virtual drainage sub-model for each drainage facility based on the drainage range and drainage direction of the same drainage facility, establish a virtual direction sub-model for the city to be tested based on each water flow direction, analyze the water basin length and water self-consumption corresponding to each virtual drainage direction in the virtual direction sub-model using the virtual drainage sub-model, and establish a virtual domain length sub-model for the city to be tested based on the water basin length.

[0109] Step 14: Based on the topographic data, establish a three-dimensional city image of the city to be tested, map the virtual drainage sub-model, the virtual direction sub-model, and the virtual domain length sub-model onto the three-dimensional city image respectively, and adjust the mapping results using the water self-consumption rate to generate the city drainage model of the city to be tested.

[0110] In this example, the water flow direction represents several water flow directions formed due to the influence of surface undulations on the water flow;

[0111] In this example, the drainage range refers to the working range of the drainage facility, and the drainage direction refers to the direction in which the drainage facility guides the water.

[0112] In this example, the watercourse length represents the length of the water flow corresponding to a virtual drainage direction;

[0113] In this example, the self-consumption of accumulated water represents the amount of accumulated water consumed under the influence of the external environment.

[0114] The working principle and beneficial effects of the above technical solution are as follows: To establish a drainage model identical to the actual situation of the city under test, the facility data of each drainage facility in the city under test is first constructed based on the facility specification data and the corresponding location of each facility. Then, the surface undulation information of the city is determined based on the topographic data of the city under test, thereby determining several water flow directions in the city under test. Further, the drainage range and drainage direction corresponding to each drainage facility are determined by combining the facility data, thereby establishing a virtual drainage sub-model corresponding to the drainage facility. Further, a virtual direction sub-model of the city under test is established based on the water flow direction. Through model adjustment and training, the water basin length and water self-consumption corresponding to each virtual drainage direction in the virtual direction sub-model are adjusted, thereby establishing a virtual domain length sub-model of the city under test. In order to improve the model, a three-dimensional city image of the city under test is established based on the topographic data. Then, the obtained virtual drainage sub-model, virtual direction sub-model and virtual domain length sub-model are all mapped onto the three-dimensional city image, and the water self-consumption measurement mapping results are used for corresponding adjustments, generating a city drainage model with the same function as the city under test, laying the foundation for subsequent precipitation training.

[0115] Example 5

[0116] Based on Example 1, the method for predicting urban surface water accumulation points based on large model technology, step 2 includes:

[0117] Step 21: Retrieve historical precipitation information of the geographical location of the city to be tested, determine several first preset precipitation intensity data based on the historical precipitation information, search for several rainstorm disaster precipitation information in the big data, determine several second preset precipitation intensity data based on the rainstorm disaster precipitation information, and establish a precipitation training database for the city to be tested based on the first precipitation intensity data and the second preset precipitation intensity data.

[0118] Step 22: Use each training data in the precipitation training database to perform time-based iterative precipitation training on the urban drainage model to obtain several training results corresponding to each training data. Based on the training results, establish the time-domain drainage information and time-domain water accumulation information of the urban drainage model under each training data.

[0119] Step 23: Determine the training precipitation intensity corresponding to each iteration cycle based on the number of iteration cycles corresponding to each training data, construct a precipitation information axis, and use the precipitation information axis corresponding to the same training data to dynamically analyze the time-domain drainage information and time-domain water accumulation information to obtain the first drainage volume and first water accumulation volume corresponding to the city under test under different time-domain precipitation intensities.

[0120] Step 24: Based on the precipitation training database, establish several dynamic precipitation intensity data. Use the dynamic precipitation intensity data to train the urban drainage model to obtain the second drainage volume and second water accumulation volume corresponding to the urban drainage model under different dynamic precipitation intensities. Based on several first drainage volumes and second drainage volumes, establish the urban drainage information of the city to be tested. Based on several first water accumulation volumes and second water accumulation volumes, establish the urban water accumulation information of the city to be tested.

[0121] In this example, the first preset precipitation intensity data represents data consistent with historical precipitation intensity;

[0122] In this example, the second preset precipitation intensity data represents the data that is consistent with the intensity of a rainstorm disaster when a rainstorm disaster occurs;

[0123] In this example, the duration iteration loop represents the process of iteratively executing the same training data in 5-minute increments.

[0124] In this example, the time-domain drainage information represents the drainage volume and drainage efficiency of the city under test when there are different durations of continuous rainfall under the same rainfall intensity.

[0125] In this example, the time-domain water accumulation information represents information such as the water accumulation volume and water accumulation efficiency of the city under test when there are different durations of continuous precipitation under the same precipitation intensity.

[0126] In this example, the precipitation information axis represents a time axis of precipitation change established based on the duration of the iteration cycle;

[0127] In this example, the first drainage volume represents the drainage volume of the city under different continuous rainfall conditions, and each iteration of the training cycle corresponds to a first drainage volume.

[0128] In this example, the first water volume represents the water volume of the city under test under different continuous rainfall conditions, and each iteration of training results corresponds to a first water volume.

[0129] In this example, dynamic precipitation data refers to precipitation data with dynamic changing characteristics established by extracting two or more training data from the precipitation training database.

[0130] In this example, the second drainage volume represents the drainage volume of the city under different dynamic precipitation intensities;

[0131] In this example, the second water volume represents the water volume of the city under different dynamic precipitation intensities.

[0132] The working principle and beneficial effects of the above technical solution are as follows: To analyze various precipitation scenarios in advance and to conduct specific analysis based on the actual precipitation in the city to be measured, the precipitation information of the city to be measured is first retrieved, thereby determining several first-preset precipitation intensity data for the city. Simultaneously, to prevent rainstorm disasters, several second-preset precipitation intensity data are established based on rainstorm disaster precipitation information, thus obtaining a precipitation training database for model training. Then, the urban drainage model is trained using the training data in the precipitation training database combined with a time-based iterative loop, thereby obtaining the time-domain drainage information and time-domain water accumulation information of the urban drainage system under different training data. Finally, the precipitation information axis of equal length is established using the number of iterations. By utilizing a precipitation information axis to dynamically analyze temporal drainage and water accumulation information, the first drainage volume and first water accumulation volume of the city under test were determined under the influence of different rainfall intensities over different durations. Since rainfall intensity can fluctuate, dynamic precipitation intensity data was established based on training data from a precipitation training database. This allowed for dynamic precipitation training of the urban drainage model, resulting in the second drainage volume and second water accumulation volume of the urban drainage model under different dynamic precipitation intensities. Through these methods, urban water accumulation information for the city under test was established. This information includes drainage and water accumulation volumes corresponding to various precipitation conditions, enabling dynamic and long-term analysis, reducing randomness, and allowing the analysis results to be better applied to daily precipitation analysis.

[0133] Example 6

[0134] Based on Example 5, the method for predicting urban surface water accumulation points based on large model technology further includes:

[0135] Based on the urban waterlogging information, several information points in the city to be tested where the urban waterlogging volume is greater than the standard waterlogging volume are determined;

[0136] Based on the urban waterlogging information, a training precipitation pattern corresponding to each information point is determined, and a warning trigger label for the information point is established based on the training precipitation pattern.

[0137] When the current precipitation pattern of the city to be tested is the same as the training precipitation pattern, the warning trigger tag is activated, a warning message is generated and displayed.

[0138] The working principle and beneficial effects of the above technical solution are as follows: warning trigger tags are added to several information points in the city under test where the water volume exceeds the standard water volume. When the current precipitation pattern of the city under test is the same as the training precipitation pattern of the information point, a corresponding warning message is generated to remind relevant personnel to respond in a timely manner.

[0139] Example 7

[0140] Based on Example 1, the method for predicting urban surface water accumulation points based on large model technology, step 3 includes:

[0141] Step 31: Establish several drainage flow paths for the city to be tested based on the urban drainage information corresponding to each rainfall intensity data, and establish several water accumulation flow paths for the city to be tested under the corresponding rainfall intensity data based on the urban water accumulation information.

[0142] Step 32: Divide the city to be tested into several urban areas according to the urban planning of the city to be tested, and obtain the first overlap information between each urban area and different drainage flow paths, and obtain the second overlap information between each urban area and different water accumulation flow paths.

[0143] Step 33: Based on the first overlap information and the second overlap information, determine the corresponding regional water accumulation of the urban area under the corresponding precipitation intensity data, and based on the first overlap information and the second overlap information, determine the water flow deposition characteristics of the corresponding urban area;

[0144] Step 34: Cities with water accumulation exceeding the preset standard water accumulation are designated as waterlogged cities. The water accumulation characteristics and the water accumulation volume are used to determine and display the water accumulation distribution information of each waterlogged city under the corresponding precipitation intensity data.

[0145] The working principle and beneficial effects of the above technical solution are as follows: To further analyze the water accumulation points and water volume of the city under test, the drainage flow path of the city is established based on the urban drainage information corresponding to each precipitation intensity data, thereby determining the water accumulation flow path of the city under test under different precipitation intensities. Then, based on the overlap information between each urban area and the drainage flow path and water accumulation flow path, the water flow deposition characteristics of the urban area are constructed, further identifying urban areas with high water accumulation. Then, using the water flow deposition characteristics and regional water accumulation, the water accumulation distribution information corresponding to each water accumulation urban area under the corresponding precipitation intensity data is determined, thereby identifying several water accumulation points in the city under test, and simultaneously determining the water accumulation volume corresponding to each water accumulation point, achieving the purpose of advance prediction.

[0146] Example 8

[0147] Based on Example 1, the method for predicting urban surface water accumulation points based on large model technology, step 4 includes:

[0148] Step 41: Using the water distribution information, establish several water accumulation points in the city under test corresponding to different precipitation intensities, and statistically analyze the water flow information corresponding to each water accumulation point;

[0149] Step 42: Draw the dynamic water flow corresponding to each urban area based on the water accumulation points corresponding to each rainfall intensity and the water flow information;

[0150] Step 43: Determine the location of the water flow collection in each corresponding urban area based on the dynamic water flow, and generate a precipitation-water accumulation point distribution map of the city to be tested.

[0151] The working principle and beneficial effects of the above technical solution are as follows: By statistically analyzing the water accumulation points and water flow information of the city under different precipitation intensities, the location of the water flow collection in the city under test is determined, thereby establishing a distribution map of water accumulation points in the city under different precipitation intensities.

[0152] Example 9

[0153] Based on Example 1, the method for predicting urban surface water accumulation points based on large model technology further includes:

[0154] The city drainage model of the city to be tested is updated based on the real-time facility construction information of the city to be tested.

[0155] The working principle and beneficial effects of the above technical solution are as follows: When construction is carried out in the city under test, the city drainage model is updated according to the real-time construction information, which reduces the lag of the model analysis results and improves the intelligence of the model.

[0156] Example 10

[0157] Based on Example 1, the method for predicting urban surface water accumulation points based on large model technology further includes:

[0158] Based on the precipitation-water accumulation point distribution map, establish and display the corresponding protection deployment plan for each precipitation intensity.

[0159] The working principle and beneficial effects of the above technical solution are as follows: by identifying high-risk areas in advance and formulating preventive measures, property losses caused by water accumulation can be effectively reduced.

[0160] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for predicting urban surface water accumulation points based on large model technology, characterized in that, include: Step 1: Collect facility data for each drainage facility in the city to be tested, and combine it with the topographic data of the city to be tested to establish an urban drainage model for the city to be tested; Step 2: Train the urban drainage model using each preset precipitation intensity data to obtain the urban drainage volume and urban water accumulation information of the city under test under different precipitation intensity data; Step 3: Analyze the regional water accumulation of each urban area in the city to be tested based on the urban drainage information and the urban water accumulation information, and determine the water accumulation distribution information of the city to be tested; Step 4: Based on the water accumulation distribution information, determine several water accumulation points corresponding to different precipitation intensities in the city to be tested, and generate a precipitation-water accumulation point distribution map of the city to be tested; Step 2 includes: Step 21: Retrieve historical precipitation information of the geographical location of the city to be tested, determine several first preset precipitation intensity data based on the historical precipitation information, search for several rainstorm disaster precipitation information in the big data, determine several second preset precipitation intensity data based on the rainstorm disaster precipitation information, and establish a precipitation training database for the city to be tested based on the first preset precipitation intensity data and the second preset precipitation intensity data. Step 22: Use each training data in the precipitation training database to perform time-based iterative precipitation training on the urban drainage model to obtain several training results corresponding to each training data. Based on the training results, establish the time-domain drainage information and time-domain water accumulation information of the urban drainage model under each training data. Step 23: Determine the training precipitation intensity corresponding to each iteration cycle based on the number of iteration cycles corresponding to each training data, construct a precipitation information axis, and use the precipitation information axis corresponding to the same training data to dynamically analyze the time-domain drainage information and time-domain water accumulation information to obtain the first drainage volume and first water accumulation volume corresponding to the city under test under different time-domain precipitation intensities. Step 24: Based on the precipitation training database, establish several dynamic precipitation intensity data, use the dynamic precipitation intensity data to train the urban drainage model to obtain the second drainage volume and second water accumulation volume corresponding to the urban drainage model under different dynamic precipitation intensities, establish the urban drainage information of the city to be tested based on several first drainage volumes and second drainage volumes, and establish the urban water accumulation information of the city to be tested based on several first water accumulation volumes and second water accumulation volumes. Also includes: Based on the urban waterlogging information, several information points in the city to be tested where the urban waterlogging volume is greater than the standard waterlogging volume are determined; Based on the urban waterlogging information, a training precipitation pattern corresponding to each information point is determined, and a warning trigger label for the information point is established based on the training precipitation pattern. When the current precipitation pattern of the city to be tested is the same as the training precipitation pattern, the warning trigger tag is activated, a warning message is generated and displayed.

2. The method for predicting urban surface water accumulation points based on large model technology as described in claim 1, characterized in that, Also includes: When precipitation occurs in the city to be tested, a real-time precipitation distribution map of the city to be tested is established based on the real-time precipitation intensity corresponding to each urban area. Find the high-frequency water accumulation points for each urban area under the corresponding real-time precipitation intensity in the precipitation-water accumulation distribution map. Using the real-time precipitation distribution map, real-time water accumulation monitoring is performed on each of the high-frequency water accumulation points to obtain the first real-time water accumulation volume corresponding to each of the high-frequency water accumulation points in the city under test. Water accumulation points are sampled for each urban area, and the real-time water accumulation is monitored for each sampled water accumulation point using the real-time precipitation distribution map to obtain the second real-time water accumulation volume corresponding to each sampled water accumulation point in the city to be tested. The real-time water accumulation information of the city to be tested is determined and displayed based on the first real-time water accumulation volume corresponding to each high-frequency water accumulation point and the second real-time water accumulation volume corresponding to each sampling water accumulation point.

3. The method for predicting urban surface water accumulation points based on large model technology as described in claim 2, characterized in that, Also includes: Based on the real-time water accumulation information, find several actual water accumulation points in the city to be tested whose real-time water accumulation is higher than the standard water accumulation; Establish a water accumulation distribution matrix for the city to be tested based on the actual distribution of water accumulation points. The difference between the real-time water accumulation and the standard water accumulation for each actual water accumulation point is input into the water accumulation distribution matrix to obtain the water accumulation information matrix of the city to be tested. Singular value decomposition is performed on the water accumulation information matrix to obtain the dimensionality-reduced data corresponding to each water accumulation difference; Based on the data trend of the dimensionality-reduced data captured by the preset linear regression model, the water accumulation trend corresponding to each actual water accumulation point is obtained. Based on the water accumulation trend, assign a corresponding number of security personnel to each actual water accumulation point.

4. The method for predicting urban surface water accumulation points based on large model technology as described in claim 1, characterized in that, Step 1 includes: Step 11: Statistically analyze the drainage facilities included in the city to be tested, obtain the facility specification data corresponding to each drainage facility, and obtain the setting location data of each drainage facility in the city to be tested. Construct facility data corresponding to each drainage facility based on the facility specification data and the setting location data. Step 12: Obtain the topographic data of the city to be tested, establish the surface undulation information of the city to be tested based on the topographic data, determine several water flow directions of the city to be tested based on the undulation information, and establish the drainage range and drainage direction corresponding to each drainage facility based on the facility data; Step 13: Construct a virtual drainage sub-model for each drainage facility based on the drainage range and drainage direction of the same drainage facility, establish a virtual direction sub-model for the city to be tested based on each water flow direction, analyze the water basin length and water self-consumption corresponding to each virtual drainage direction in the virtual direction sub-model using the virtual drainage sub-model, and establish a virtual domain length sub-model for the city to be tested based on the water basin length. Step 14: Based on the topographic data, establish a three-dimensional city image of the city to be tested, map the virtual drainage sub-model, the virtual direction sub-model, and the virtual domain length sub-model onto the three-dimensional city image respectively, and adjust the mapping results using the water self-consumption rate to generate the city drainage model of the city to be tested.

5. The method for predicting urban surface water accumulation points based on large model technology as described in claim 1, characterized in that, Step 3 includes: Step 31: Establish several drainage flow paths for the city to be tested based on the urban drainage information corresponding to each rainfall intensity data, and establish several water accumulation flow paths for the city to be tested under the corresponding rainfall intensity data based on the urban water accumulation information. Step 32: Divide the city to be tested into several urban areas according to the urban planning of the city to be tested, and obtain the first overlap information between each urban area and different drainage flow paths, and obtain the second overlap information between each urban area and different water accumulation flow paths. Step 33: Based on the first overlap information and the second overlap information, determine the corresponding regional water accumulation of the urban area under the corresponding precipitation intensity data, and based on the first overlap information and the second overlap information, determine the water flow deposition characteristics of the corresponding urban area; Step 34: Cities with water accumulation exceeding the preset standard water accumulation are designated as waterlogged cities. The water accumulation characteristics and the water accumulation volume are used to determine and display the water accumulation distribution information of each waterlogged city under the corresponding precipitation intensity data.

6. The method for predicting urban surface water accumulation points based on large model technology as described in claim 1, characterized in that, Step 4 includes: Step 41: Using the water distribution information, establish several water accumulation points in the city under test corresponding to different precipitation intensities, and statistically analyze the water flow information corresponding to each water accumulation point; Step 42: Draw the dynamic water flow corresponding to each urban area based on the water accumulation points corresponding to each rainfall intensity and the water flow information; Step 43: Determine the location of the water flow collection in each corresponding urban area based on the dynamic water flow, and generate a precipitation-water accumulation point distribution map of the city to be tested.

7. The method for predicting urban surface water accumulation points based on large model technology as described in claim 1, characterized in that, Also includes: The city drainage model of the city to be tested is updated based on the real-time facility construction information of the city to be tested.

8. The method for predicting urban surface water accumulation points based on large model technology as described in claim 1, characterized in that, Also includes: Based on the precipitation-water accumulation point distribution map, establish and display the corresponding protection deployment plan for each precipitation intensity.

Images