Urban traffic evaluation method and device based on spatial overlay analysis

Abstract

To provide an urban traffic evaluation method and device based on spatial overlay analysis that improves flood disaster prediction precision so as to realize accurate evaluation of urban traffic.SOLUTION: A method comprises the steps of: carrying out spatial data overlay on station monitoring rainfall data, urban surface data and urban underground pipeline data, and generating spatial overlay data for each urban grid; training a predetermined deep learning model to generate an urban flood level prediction model; obtaining weather forecast data in a target urban area, and determining a flood level prediction result for each grid by using the urban flood level prediction model; collecting urban population distribution data in a current time period, inputting the urban population distribution data into an urban traffic flow prediction model, and generating a traffic flow prediction result for each grid; and performing spatial overlay analysis on the flood level prediction result for each grid and the traffic flow prediction result for each grid, and evaluating urban traffic in the urban grid.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to the technical field of urban flood disaster and road traffic safety, in particular to a method and device for urban traffic assessment based on spatial overlay analysis. [Background technology]

[0002] Due to the suddenness of flooding, the uncertainty of the time and scope of occurrence, and the influence of factors such as the city's surface, topography, underground, and rainfall type, it is very difficult to take measures against flooding. Most of the conventional methods for predicting heavy rain inundation in cities make short-term predictions of areas where flooding may occur in the city based on real-time information from water level monitoring stations in the city. Due to insufficient consideration of influencing factors and difficulty in determining the relationship between multiple influencing factors, flood predictions are not accurate enough. Therefore, how to improve the accuracy of flood prediction and realize accurate evaluation of urban traffic is currently a technical problem that is urgently needed to be solved. Summary of the Invention [Problem to be solved by the invention]

[0003] In view of the above circumstances, the present invention provides a method and apparatus for evaluating urban traffic based on spatial overlay analysis to solve the technical problem of how to improve the accuracy of flood prediction and realize accurate evaluation of urban traffic. [Means for solving the problem]

[0004] In a first aspect, the present invention provides an urban traffic evaluation method based on spatial overlay analysis, including the steps of: acquiring station monitoring rainfall data, urban surface data, and urban underground pipeline data in a target urban area, and performing spatial data overlay on the station monitoring rainfall data, urban surface data, and urban underground pipeline data in the target urban area to generate spatial overlay data for each urban grid; training a predetermined deep learning model based on the spatial overlay data for each urban grid, and generating an urban flood inundation depth prediction model; acquiring weather forecast data in the target urban area, and determining a flood depth prediction result for each grid using the urban flood inundation depth prediction model based on the weather forecast data; collecting urban population distribution data in a current period, inputting the urban population distribution data in the current period into the urban traffic flow prediction model, and generating a traffic flow prediction result for each grid; performing spatial overlay analysis on the flood depth prediction result for each grid and the traffic flow prediction result for each grid, and evaluating the urban traffic in the city grid based on the spatial overlay analysis result, and generating an urban traffic evaluation result.

[0005] The urban traffic assessment method based on spatial overlay analysis in this embodiment is based on the idea of ​​spatial overlay analysis and the idea of ​​urban spatial domain discretization, and overlays all factors affecting urban flood inundation depth, such as sky, surface, underground, etc., onto the same urban grid, generates spatial overlay data for each urban grid, trains and generates an urban flood inundation depth prediction model based on the generated spatial overlay data for each urban grid, and further uses the urban flood inundation depth prediction model to determine the inundation depth prediction result for each grid based on weather forecast data, thereby realizing the prediction of the inundation depth of different areas of the city in combination with the weather forecast data, and improving the prediction accuracy of flood damage. Finally, generates traffic flow prediction result for each grid according to population distribution data and urban traffic flow prediction model, predicts the spatial distribution of traffic flow at different time nodes in the future of the city's traffic, and performs spatial overlay analysis together with the inundation depth prediction result for each urban grid, thereby realizing an accurate evaluation of urban traffic.

[0006] In one optional embodiment, the step of performing spatial data overlay on the station monitoring rainfall data, urban surface data, and urban underground pipeline data in the target urban area and generating spatial overlay data for each urban grid includes the steps of performing spatial discretization processing on the target urban area and generating multiple urban grids; and dividing the station monitoring rainfall data, urban surface data, and urban underground pipeline data into multiple urban grids respectively according to spatial topological relationships, and generating spatial overlay data for each urban grid.

[0007] The urban traffic assessment method based on spatial overlay analysis in this embodiment is based on the idea of ​​spatial overlay analysis and the idea of ​​urban spatial domain discretization, and overlays all elements such as sky, surface, underground, etc. that affect the flood inundation depth of the city onto the same grid, with each unit grid representing the location, area attributes and precipitation information of the city, thereby enabling accurate prediction of specific areas where urban floods will occur and comprehensive and detailed prediction of the distribution of floods in different areas of the city.

[0008] In one optional embodiment, the step of training a predetermined deep learning model based on the spatial overlay data for each city grid and generating an urban flood inundation depth prediction model includes the steps of inputting the spatial overlay data for each city grid into a predetermined flood inundation deep learning model and generating an output result of the predetermined flood inundation deep learning model, where one predetermined flood inundation deep learning model is constructed for each city grid; and the steps of obtaining urban flood inundation distribution data, comparing the output result of the predetermined flood inundation deep learning model with the urban flood inundation distribution data, and generating an urban flood inundation depth prediction model if the comparison result satisfies a predetermined condition.

[0009] The urban traffic assessment method based on spatial overlay analysis in this embodiment realizes learning of the associations among multiple influencing factors by constructing an urban flood inundation depth prediction model, avoids inaccurate prediction of urban flood inundation depth caused by the uncertainty of the associations among multiple influencing factors, and thereby improves the prediction accuracy of urban flood inundation depth.

[0010] In one alternative embodiment, the step of determining flood depth forecast results for each grid using an urban flood inundation depth forecast model based on weather forecast data includes the steps of inputting a weather forecast dataset for the urban target area into a station precipitation forecast model and outputting station precipitation forecast data, and inputting the station precipitation forecast data into the urban flood inundation depth forecast model and outputting flood depth forecast results for each grid.

[0011] The urban traffic assessment method based on spatial overlay analysis in this embodiment combines grid flood depth prediction and weather forecast data to realize prediction of urban flood spatial distribution in different time periods such as hourly, daily, and weekly future periods, and realize advance arrangement of traffic transportation, thereby maximally ensuring the safety of residents' property and people, and preventing major damage to urban development.

[0012] In one alternative embodiment, the urban flood inundation depth prediction model is a long-short-term memory deep learning model.

[0013] In the urban transportation assessment method based on spatial overlay analysis of this embodiment, the long-short-term memory deep learning model memorizes the training results in the previous period, and further adjusts the model based on the input data of the current time series. As the input data continues to increase, the simulation results of the model continue to be corrected, which can better allocate the information of historical units and has the ability to capture long-term dependencies in the time series.

[0014] In one optional embodiment, the step of performing a spatial overlay analysis on the flood depth prediction result for each grid and the traffic flow prediction result for each grid, evaluating urban traffic in the city grid based on the spatial overlay analysis result, and generating an urban traffic evaluation result includes a step of comparing the flood depth prediction result for each grid with a predetermined depth threshold and generating a flood depth level, a step of comparing the traffic flow prediction result for each grid with a predetermined flow threshold and generating a traffic flow level, and a step of determining an urban traffic hazard level in the city grid based on the flood depth level and the traffic flow level, and taking the urban traffic hazard level as the urban traffic evaluation result.

[0015] The urban traffic assessment method based on spatial overlay analysis in this embodiment obtains the danger levels of different areas of the city based on the flood depth level and the traffic flow level, and pre-arranges and directs the urban road traffic based on the predicted danger level, thereby maximally ensuring the property and personal safety of residents and preventing major damage to the development of the city.

[0016] In one optional embodiment, the method further includes real-time urban road traffic dispatch and early risk warning based on the spatial overlay analysis result.

[0017] The urban traffic assessment method based on spatial overlay analysis according to the present embodiment can maximize the protection of residents' property and personal safety and prevent major damage to the development of the city.

[0018] In a second aspect, the present invention provides a first overlay module, which is used for obtaining station monitoring rainfall data, urban surface data, and urban underground pipeline data in a target urban area, and performing spatial data overlay on the station monitoring rainfall data, urban surface data, and urban underground pipeline data in the target urban area to generate spatial overlay data for each urban grid; a first training module, which is used for training a predetermined deep learning model based on the spatial overlay data for each urban grid, and generating an urban flood inundation depth prediction model; and a second training module, which is used for obtaining meteorological forecast data in the target urban area, and performing spatial overlay on the station monitoring rainfall data, urban surface data, and urban underground pipeline data in the target urban area to generate spatial overlay data for each urban grid. The present invention provides an urban traffic evaluation device based on spatial overlay analysis, including a first acquisition module used for determining a flood depth prediction result for each grid using a flood inundation depth prediction model, a first collection module used for collecting urban population distribution data in a current period, inputting the urban population distribution data in the current period into an urban traffic flow prediction model, and generating a traffic flow prediction result for each grid, and a first evaluation module used for performing a spatial overlay analysis on the flood depth prediction result for each grid and the traffic flow prediction result for each grid, evaluating the urban traffic in the city grid based on the spatial overlay analysis result, and generating an urban traffic evaluation result.

[0019] In a third aspect, the present invention provides a computer device comprising: a memory; and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, the processor executing the computer instructions to perform the method for urban traffic assessment based on spatial overlay analysis as set forth in the first aspect above or any corresponding embodiment thereof.

[0020] In a fourth aspect, the present invention provides a computer-readable storage medium having stored thereon computer instructions for causing a computer to execute the method for urban traffic assessment based on spatial overlay analysis as set forth in the first aspect or any corresponding embodiment thereof. [Brief description of the drawings]

[0021] In order to more clearly describe the specific embodiments of the present invention or the technical solutions of the prior art, the drawings that need to be used to describe the specific embodiments or the prior art will be briefly described below. It will be apparent that the drawings described below are some embodiments of the present invention, and those skilled in the art can obtain other drawings based on these drawings without creative labor.

[0022] [Figure 1] 1 is a flowchart (1) of a method for urban transportation assessment based on spatial overlay analysis according to an embodiment of the present invention. [Diagram 2] FIG. 1 is a schematic diagram (1) of a backpropagation neural network model according to an embodiment of the present invention. [Diagram 3] 2 is a flowchart (2) of an urban transportation assessment method based on spatial overlay analysis according to an embodiment of the present invention. [Figure 4] 3 is a flowchart (3) of an urban transportation assessment method based on spatial overlay analysis according to an embodiment of the present invention. [Diagram 5] FIG. 2 is a unit structure diagram of a long-short-term memory deep learning model according to an embodiment of the present invention. [Figure 6] 4 is a flowchart (4) of a method for urban transportation assessment based on spatial overlay analysis according to an embodiment of the present invention. [Figure 7] FIG. 2 is a schematic diagram (2) of a backpropagation neural network model according to an embodiment of the present invention. [Figure 8] 5 is a flowchart (5) of a method for urban transportation assessment based on spatial overlay analysis according to an embodiment of the present invention. [Figure 9] 6 is a flowchart (6) of a method for urban transportation assessment based on spatial overlay analysis according to an embodiment of the present invention. [Figure 10] FIG. 1 is a structural block diagram of an urban traffic assessment device based on spatial overlay analysis according to an embodiment of the present invention; [Figure 11] FIG. 2 is a schematic diagram of the hardware structure of a computer device according to an embodiment of the present invention; DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0023] In order to make the objectives, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the drawings of the embodiments of the present invention, and it will be apparent that the described embodiments are only some of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments that a person skilled in the art can obtain without creative labor fall within the scope of protection of the present invention.

[0024] The urban traffic assessment method based on spatial overlay analysis according to the present specification can be applied to an urban flood prediction electronic device and a road traffic assessment electronic device, which may include, but is not limited to, a notebook computer, a desktop computer, and a mobile terminal such as a mobile phone or a tablet computer, etc. Of course, the urban traffic assessment method based on spatial overlay analysis according to the present specification can also be applied to an application program executed on the above electronic devices.

[0025] The embodiment of the present invention provides an urban traffic assessment method based on spatial overlay analysis, and by using the research method of spatial element overlay analysis, the accuracy of flood prediction is improved and the accurate assessment of urban traffic is realized.

[0026] In an embodiment of the present invention, an embodiment of an urban traffic assessment method based on spatial overlay analysis is provided, and it is noted that the steps illustrated in the flowcharts of the drawings may be performed in a computer system, such as a set of computer-executable instructions, and that although a logical order is shown in the flowcharts, in some cases the steps shown or described may be performed in an order different from that shown here.

[0027] In this embodiment, a method for urban traffic assessment based on spatial overlay analysis is provided, which can be applied to the above-mentioned notebook computers, desktop computers, mobile terminals, and mobile terminals such as mobile phones and tablet computers. FIG. 1 is a flowchart of the method for urban traffic assessment based on spatial overlay analysis according to an embodiment of the present invention. As shown in FIG. 1, the process includes the following steps:

[0028] Step S101: obtain station monitoring rainfall data, urban surface data and urban underground pipeline data within a target urban area, and perform spatial data overlay on the station monitoring rainfall data, urban surface data and urban underground pipeline data within the target urban area to generate spatial overlay data for each urban grid.

[0029] Specifically, as shown in Table 1, the data affecting urban flooding includes rainfall data, urban surface data, and urban underground pipeline data.

[0030] Table 1 JPEG2025026360000002.jpg194170

[0031] Rainfall data is a determining factor of flood disasters. Without precipitation, it is difficult for urban floods to occur (except coastal cities). Different precipitation distributions in urban precipitation (based on ground monitoring stations or radar rainfall monitoring) represent different precipitation spatial forms (rainfall intensity, rainfall duration, rainfall type, rainfall center, rainfall trajectory) occurring above the city, and different rainfall spatial distributions directly determine the occurrence of flood disasters in a city.

[0032] In addition to precipitation data, important factors to be considered are the different surface conditions of the urban surface (roads, forest land, grassland, residential areas) and the changes in the surface topography (ground gradient direction, topography). Conventional urban flood predictions focused on precipitation factors and ignored the influence of the surface factors of the city itself. Therefore, although the conventional research methods have a relatively high prediction accuracy for the possibility of flood occurrence based on precipitation data, they ignore the influence of the urban surface factors, so it is generally difficult to accurately predict the specific areas of the city where flooding will occur. In contrast, the present embodiment takes into account the spatial distribution situation of the urban surface data to realize an accurate prediction of the specific areas of the city where flooding will occur.

[0033] The urban underground pipeline data includes underground drainage pipeline outlets, drainage pipeline diameters, and sewage discharge capacities, which can depict the response capabilities of different areas of the city in the event of flooding, and further provide detailed predictions of the possibility of flooding in different areas of the city.

[0034] Urban flooding is analyzed by considering the spatial distribution of the main urban flood affecting elements in the sky, surface and underground of the city, and conducting overlay analysis taking into account all spatial distribution scenes.

[0035] Step S102: Train a predetermined deep learning model based on the spatial overlay data for each urban grid to generate an urban flood inundation depth prediction model.

[0036] Step S103: Obtain weather forecast data within the target urban area, and determine the flood depth prediction result for each grid using the urban flood inundation depth prediction model based on the weather forecast data.

[0037] Step S104: Collect urban population distribution data in the current period, input the urban population distribution data in the current period into the urban traffic flow prediction model, and generate traffic flow prediction results for each grid.

[0038] Specifically, historical population distribution data and historical road traffic flow data of the target urban area in a predetermined historical period are obtained, the historical population distribution data in the predetermined historical period is input into a predetermined road traffic flow prediction model, an output result of the predetermined road traffic flow prediction model is obtained, the output result of the predetermined road traffic flow prediction model is compared with the historical road traffic flow data in the predetermined historical period, and if the comparison result satisfies a predetermined condition, an urban road traffic flow prediction model is generated.

[0039] In addition, the real-time road traffic flow of a city is related to its population distribution (residential population distribution), the time of day, the number of cars and light vehicles in the city, etc., but for a city, its road traffic flow is closely related to the date of the year and the time of the day. For example, in the winter of the year, the temperature is low, so the number of light vehicles decreases, and more residents choose public transportation when going out. On the other hand, as the temperature rises, the corresponding number gradually increases. In addition, for example, on the day before holidays such as May Day and National Day, many people travel by car, so the number of cars increases sharply throughout the day. In a day, 8-9 am is the morning rush hour, and 5-6 pm is the evening rush hour. The traffic flow in these two time periods increases sharply. According to the different population distribution laws of the city, the traffic flow in different areas of the corresponding city is also ... Traffic flow has a certain rule, and when considered comprehensively, a back-propagation (BP) neural network model can be adopted to predict the traffic flow in different areas of a city. As shown in Figure 2, the target city is divided into multiple city grids based on the grid division form, and a BP neural network is built for each grid into which the city is divided. The city's date and time (month, day, hour) (X1) for one year and the number of residents in the grid (X2) are used as input information, and the traffic flow corresponding to the grid is used as output (Y) to train the BP neural network, so that a traffic flow prediction result at any time in a year can be obtained. The model is trained based on traffic flow information in recent years (for example, the most recent five years) to generate an urban road traffic flow prediction model.

[0040] Step S105: A spatial overlay analysis is performed on the flood depth prediction result for each grid and the traffic flow prediction result for each grid, and the urban traffic in the city grid is evaluated based on the spatial overlay analysis result, and an urban traffic evaluation result is generated.

[0041] Specifically, urban road traffic transportation will be arranged in real time based on the results of spatial overlay analysis, early danger warnings will be issued, and the safety of residents' property and personal life will be ensured to the maximum extent possible, preventing major damage to urban development.

[0042] Furthermore, based on the constructed urban flood inundation depth prediction model and urban traffic flow prediction model for different grids, a spatial overlay analysis needs to be further performed in order to timely adjust urban traffic, that is, a spatial overlay analysis is further performed on the flood depth prediction result for each grid and the traffic flow prediction result for each grid, and for the same urban area, when the flood depth prediction result is greater than a certain danger threshold and the predicted traffic flow of the area reaches a certain value, the urban area is a danger area, and the traffic danger of different areas can be warned based on the overlay relationship between the flood depth and the urban road traffic flow.

[0043] In addition, based on the urban flood inundation depth prediction model, it timely predicts areas that may cause serious danger, predicts the specific areas where urban flooding will occur in the future "hours-days-weeks", disperses traffic within the "unit grid", and feeds back the prediction results to the traffic command department, blocking road sections several hours in advance and relocating vehicles and personnel.

[0044] In addition, since there is still a large degree of uncertainty in weather changes and predictions, when heavy rain is about to occur, the model will interact with meteorological data in real time, and thereby continue to revise the improved results of the model, and then link with the transportation department in real time to revise the traffic command strategy, so as to ensure optimal road traffic conditions every time heavy rain occurs in the city, thereby minimizing casualties and property losses of people caused by urban floods, and achieving timely adjustment, optimization and layout of urban transportation.

[0045] Furthermore, based on the danger level of the spatial overlay analysis results, responses are made in order from high to low. For extremely dangerous areas at the highest level, personnel in the current area should be notified to evacuate as soon as possible, traffic flow should be quickly relocated, and traffic control should be implemented to prevent personnel from entering, thereby downgrading the danger level. For non-dangerous areas at the lowest level, no management is required. However, in areas with high traffic flow and deep flooding, there is generally a possibility of hidden danger, so if there is still capacity in traffic command, management can be expanded to non-key areas, thereby enabling a more timely response to the emergency.

[0046] The urban traffic assessment method based on spatial overlay analysis in this embodiment is based on the idea of ​​spatial overlay analysis and the idea of ​​urban spatial domain discretization, and overlays all factors affecting urban flood inundation depth, such as sky, surface, underground, etc., onto the same urban grid, generates spatial overlay data for each urban grid, trains and generates an urban flood inundation depth prediction model based on the generated spatial overlay data for each urban grid, and further uses the urban flood inundation depth prediction model to determine the inundation depth prediction result for each grid based on weather forecast data, thereby realizing the prediction of the inundation depth of different areas of the city in combination with the weather forecast data, and improving the prediction accuracy of flood damage. Finally, generates traffic flow prediction result for each grid according to population distribution data and urban traffic flow prediction model, predicts the spatial distribution of traffic flow at different time nodes in the future of the city's traffic, and performs spatial overlay analysis together with the inundation depth prediction result for each urban grid, thereby realizing an accurate evaluation of urban traffic.

[0047] In this embodiment, a method for urban traffic assessment based on spatial overlay analysis is provided, which can be applied to the above-mentioned notebook computers, desktop computers, mobile terminals, and mobile terminals such as mobile phones and tablet computers. FIG. 3 is a flowchart of the method for urban traffic assessment based on spatial overlay analysis according to an embodiment of the present invention. As shown in FIG. 3, the process includes the following steps:

[0048] Step S301: Obtain station monitoring rainfall data, urban surface data and urban underground pipeline data within a target urban area, and perform spatial data overlay on the station monitoring rainfall data, urban surface data and urban underground pipeline data within the target urban area to generate spatial overlay data for each urban grid.

[0049] Specifically, the above step S301 includes steps S3011 to S3012.

[0050] Step S3011: Perform spatial discretization processing on the target urban area to generate multiple urban grids.

[0051] Specifically, precipitation data is a time- and space-dual-varying element, while urban surface data and urban underground pipeline data are spatially-varying elements. For a certain precipitation, whether flooding will occur in a city, the location of flooding and the extent of flooding depend on the spatial distribution of precipitation intensity, the spatial characteristic distribution of the urban surface, and the spatial distribution of the urban underground drainage characteristics. In order to take into account the spatial overlay of all influencing elements, this embodiment divides the grid into rectangular grids such as squares (e.g., 10m*10m) in the form of grid division, thereby performing spatial discretization for the city, and the spatial discretization process of the city provides a basis for subsequent regional prediction of flood risk in different areas of the city.

[0052] Step S3012: According to the spatial topology relationship, the station monitoring rainfall data, the city surface data and the city underground pipeline data are respectively divided into multiple city grids, and spatial overlay data for each city grid is generated.

[0053] Specifically, based on each divided rectangular grid, all information such as precipitation intensity data, precipitation duration data, urban surface land use type, urban slope, urban underground drainage pipeline width, drainage capacity, and water retention capacity are plotted on all grids in the city, all influencing element attributes are assigned to the grids, and the spatial topology of all elements that determine whether flooding will occur in the grid is overlaid on the grid, thereby realizing the spatial overlay processing of all influencing elements in space, where each unit grid represents the urban area position, area attributes, and precipitation information, and each grid contains the information after spatial overlay of the influencing elements in Table 1.

[0054] Step S302: Train a predetermined deep learning model based on the spatial overlay data for each city grid, and generate an urban flood inundation depth prediction model. For details, please refer to step S102 of the embodiment shown in Figure 1, and redundant explanations will be omitted here.

[0055] Step S303: Obtain weather forecast data within the target urban area, and based on the weather forecast data, use the urban flood inundation depth prediction model to determine the flood depth prediction result for each grid. For details, please refer to step S103 of the embodiment shown in Figure 1, and duplicate explanations will be omitted here.

[0056] Step S304: Collect urban population distribution data in the current period, input the urban population distribution data in the current period into the urban traffic flow prediction model, and generate traffic flow prediction results for each grid. For details, refer to step S104 in the embodiment shown in Figure 1, and redundant explanations will be omitted here.

[0057] Step S305: A spatial overlay analysis is performed on the flood depth prediction results for each grid and the traffic flow prediction results for each grid. Based on the spatial overlay analysis results, urban traffic within the city grid is evaluated and urban traffic evaluation results are generated. For details, refer to step S105 of the embodiment shown in Figure 1, and redundant explanations will be omitted here.

[0058] The urban traffic assessment method based on spatial overlay analysis in this embodiment is based on the idea of ​​spatial overlay analysis and the idea of ​​urban spatial domain discretization, and overlays all elements such as sky, surface, underground, etc. that affect the flood inundation depth of the city onto the same grid, with each unit grid representing the location, area attributes and precipitation information of the city, thereby enabling accurate prediction of specific areas where urban floods will occur and comprehensive and detailed prediction of the distribution of floods in different areas of the city.

[0059] In this embodiment, a method for urban traffic assessment based on spatial overlay analysis is provided, which can be applied to the above-mentioned notebook computers, desktop computers, mobile terminals, and mobile terminals such as mobile phones and tablet computers. Figure 4 is a flowchart of the method for urban traffic assessment based on spatial overlay analysis according to an embodiment of the present invention. As shown in Figure 4, the process includes the following steps:

[0060] Step S401: Obtain station monitoring rainfall data, urban surface data and urban underground pipeline data within a target urban area, and perform spatial data overlay on the station monitoring rainfall data, urban surface data and urban underground pipeline data within the target urban area to generate spatial overlay data for each urban grid. For details, please refer to step S101 of the embodiment shown in Figure 1, and redundant explanations will be omitted here.

[0061] Step S402: Train a predetermined deep learning model based on the spatial overlay data for each urban grid to generate an urban flood inundation depth prediction model.

[0062] Specifically, the above step S402 includes steps S4021 to S4022.

[0063] Step S4021: The spatial overlay data for each city grid is input into a predetermined flood inundation deep learning model, and an output result of the predetermined flood inundation deep learning model is generated, and one predetermined flood inundation deep learning model is constructed for each city grid.

[0064] Specifically, the predetermined flood inundation deep learning model may be a long-short-term memory deep learning model (LSTM).

[0065] Step S4022: Urban flood inundation distribution data is acquired, and the output result of a predetermined flood inundation deep learning model is compared with the urban flood inundation distribution data. If the comparison result satisfies a predetermined condition, an urban flood inundation depth prediction model is generated.

[0066] Specifically, the urban flood inundation depth prediction model is a long-short-term memory deep learning model, and a spatial overlay data format is constructed for each urban grid, so that the occurrence of urban flooding can be predicted. In this embodiment, the LSTM deep learning model is used to overlay spatial data of three aspects, namely, sky, surface, and underground, onto each grid unit of the city, thereby extracting scene data of urban flooding that has occurred in the past in the unit, and constructing one LSTM model for each constructed urban unit grid. The spatial trinity information (sky, surface, underground) of each constructed unit is used as input data for the LSTM deep learning model of each unit grid, and the corresponding monitored historical urban flood inundation distribution data (city By using the data (city flood distribution area and flood depth data) as calibration data for each unit grid, the LSTM model is trained and predictive simulation results are obtained. The reason for selecting the LSTM deep learning neural network model in this embodiment is that the model is a time series model, that is, input data is input in order of occurrence time and trained, the LSTM model remembers the training results in the previous period (time t-1), and further adjusts the model based on the input data of the current time series (time t). As the input data continues to increase, the simulation results of the model continue to be revised, which can better allocate historical unit information and has the ability to capture long-term dependencies in time series.

[0067] Furthermore, the structure diagram of the LSTM model is shown in Figure 5, where the LSTM includes multiple time repetition units, each of which has one additional memory state c t , and a number of gates for controlling information flow, including an input gate, a memory gate, a forget gate and an output gate. In FIG. 5, h t represents the implicit state, and x t represents the input at time t, and f t , i t ,o t , and c t represent the forget gate, input gate, output gate and cell state at time t, respectively, and the related calculation method of LSTM is as follows:

[0068] JPEG2025026360000003.jpg124170

[0069] JPEG2025026360000004.jpg61170

[0070] In addition, for a city, the changes in the surface characteristic conditions and underground drainage characteristic conditions of the city are generally relatively small, or relatively small within a short period of time, so during simulation, the model can be trained by simply continuing to modify the rainfall data set. However, when considering the long-term situation, the surface and underground of a city may undergo significant changes, which is difficult to depict with traditional circulation models such as neural networks, because it is difficult to consider time-series changes. On the other hand, the LSTM model can fully consider such changes, and by modifying the surface characteristic data or underground characteristic data based on the data set at the current time, the model can continue the previous training content and adjust the originally trained model together with the new input data, thereby realizing a prediction model that "self-adjusts" and continues to learn by itself according to a time-series input, and can accurately predict urban floods.

[0071] By continuing to train the model, an LSTM urban flood inundation model for each grid of the city can be obtained, and the output result of the model is the magnitude of flood inundation depth H that will occur in that grid unit under the input of precipitation, where 0 indicates no inundation, and greater than 0 indicates inundation depth will occur; the output result of the predetermined flood inundation deep learning model is compared with the urban flood inundation distribution data; if the comparison result meets a predetermined condition, an urban flood inundation depth prediction model is generated, where the predetermined condition may be that the loss value is smaller than a predetermined threshold, and those skilled in the art can set the magnitude of the predetermined threshold according to actual needs, and there is no limitation here; the flood inundation depth output by each grid of the city can be predicted based on the urban flood inundation depth prediction model, and the flood inundation distribution situation of the entire city can be obtained.

[0072] Step S403: Obtain weather forecast data within the target urban area, and based on the weather forecast data, use the urban flood inundation depth prediction model to determine the flood depth prediction result for each grid. For details, please refer to step S103 of the embodiment shown in Figure 1, and redundant explanations will be omitted here.

[0073] Step S404: Collect urban population distribution data in the current period, input the urban population distribution data in the current period into the urban traffic flow prediction model, and generate traffic flow prediction results for each grid. For details, refer to step S104 in the embodiment shown in Figure 1, and redundant explanations will be omitted here.

[0074] Step S405: A spatial overlay analysis is performed on the flood depth prediction results for each grid and the traffic flow prediction results for each grid. Based on the spatial overlay analysis results, urban traffic within the city grid is evaluated and urban traffic evaluation results are generated. For details, refer to step S105 of the embodiment shown in Figure 1, and redundant explanations will be omitted here.

[0075] The urban traffic assessment method based on spatial overlay analysis in this embodiment realizes learning of the associations among multiple influencing factors by constructing an urban flood inundation depth prediction model, avoids inaccurate prediction of urban flood inundation depth caused by the uncertainty of the associations among multiple influencing factors, and thereby improves the prediction accuracy of urban flood inundation depth.

[0076] In this embodiment, a method for urban traffic assessment based on spatial overlay analysis is provided, which can be applied to the above-mentioned notebook computers, desktop computers, mobile terminals, and mobile terminals such as mobile phones and tablet computers. Figure 6 is a flowchart of the method for urban traffic assessment based on spatial overlay analysis according to an embodiment of the present invention. As shown in Figure 6, the process includes the following steps:

[0077] Step S601: Obtain station monitoring rainfall data, urban surface data and urban underground pipeline data within a target urban area, and perform spatial data overlay on the station monitoring rainfall data, urban surface data and urban underground pipeline data within the target urban area to generate spatial overlay data for each urban grid. For details, please refer to step S101 of the embodiment shown in Figure 1, and redundant explanations will be omitted here.

[0078] Step S602: Train a predetermined deep learning model based on the spatial overlay data for each city grid, and generate an urban flood inundation depth prediction model. For details, please refer to step S102 of the embodiment shown in Figure 1, and redundant explanations will be omitted here.

[0079] Step S603: Obtain weather forecast data within the target urban area, and determine the flood depth prediction result for each grid using the urban flood inundation depth prediction model based on the weather forecast data.

[0080] Specifically, the above step S603 includes steps S6031 to S6032.

[0081] Step S6031: The weather forecast dataset of the urban target area is input into the station precipitation forecast model, and the station precipitation forecast data is output.

[0082] Specifically, after constructing an urban flood inundation depth prediction model, it is applied to predict whether floods will occur in the city in the future and the areas where they will occur. The input of the urban flood inundation depth prediction model constructed in this embodiment is the actual precipitation monitoring data of different stations distributed in the city. Advance prediction requires access to station precipitation forecast data. Since station precipitation forecast data is highly related to weather forecast data, this embodiment is taken as an example to realize estimating future station monitoring data based on future weather forecast data. A station precipitation forecast model is constructed to associate the weather forecast data with the station precipitation forecast data, thereby predicting station precipitation data. Taking the station precipitation forecast model as an example, the BP neural network model for each grid is shown in Figure 7, where the input data is weather data (F) and the output precipitation data for each grid is (P).

[0083] Furthermore, historical weather data and station historical monitoring rainfall data within the target urban area are obtained, the historical weather data is input into the station precipitation forecast model to be trained, the output result of the station precipitation forecast model to be trained is obtained, the output result of the station precipitation forecast model to be trained is compared with the station historical monitoring rainfall data, and if the comparison result meets a predetermined condition, a station precipitation forecast model is generated.

[0084] Step S6032: The station precipitation forecast data is input to the urban flood inundation depth forecast model, and the inundation depth forecast results for each grid are output.

[0085] Specifically, future heavy rain flooding scenes in cities can be predicted based on the station precipitation prediction model and weather forecast data, and flood depth prediction results for each grid can be output. The predicted future scenes in which flooding will occur in cities correspond one-to-one to scale data such as hourly, daily, and weekly weather forecasts. That is, in this embodiment, the urban flood forecast period coincides with the weather forecast period.

[0086] Step S604: Collect urban population distribution data in the current period, input the urban population distribution data in the current period into the urban traffic flow prediction model, and generate traffic flow prediction results for each grid. For details, refer to step S104 in the embodiment shown in Figure 1, and redundant explanations will be omitted here.

[0087] Step S605: A spatial overlay analysis is performed on the flood depth prediction results for each grid and the traffic flow prediction results for each grid, and urban traffic within the city grid is evaluated based on the spatial overlay analysis results, and urban traffic evaluation results are generated. For details, refer to step S105 of the embodiment shown in Figure 1, and redundant explanations will be omitted here.

[0088] The urban traffic assessment method based on spatial overlay analysis in this embodiment combines grid flood depth prediction and weather forecast data to realize prediction of urban flood spatial distribution in different time periods such as hourly, daily, and weekly future periods, and realize advance arrangement of traffic transportation, thereby maximally ensuring the safety of residents' property and people, and preventing major damage to urban development.

[0089] In this embodiment, a method for urban traffic assessment based on spatial overlay analysis is provided, which can be applied to the above-mentioned notebook computers, desktop computers, mobile terminals, and mobile terminals such as mobile phones and tablet computers. Figure 8 is a flowchart of the method for urban traffic assessment based on spatial overlay analysis according to an embodiment of the present invention. As shown in Figure 8, the process includes the following steps:

[0090] Step S801: Obtain station monitoring rainfall data, urban surface data and urban underground pipeline data within a target urban area, and perform spatial data overlay on the station monitoring rainfall data, urban surface data and urban underground pipeline data within the target urban area to generate spatial overlay data for each urban grid. For details, please refer to step S101 of the embodiment shown in Figure 1, and redundant explanations will be omitted here.

[0091] Step S802: Train a predetermined deep learning model based on the spatial overlay data for each city grid, and generate an urban flood inundation depth prediction model. For details, please refer to step S102 of the embodiment shown in Figure 1, and redundant explanations will be omitted here.

[0092] Step S803: Obtain weather forecast data within the target urban area, and based on the weather forecast data, use the urban flood inundation depth prediction model to determine the flood depth prediction result for each grid. For details, please refer to step S103 of the embodiment shown in Figure 1, and redundant explanations will be omitted here.

[0093] Step S804: Collect urban population distribution data in the current period, input the urban population distribution data in the current period into the urban traffic flow prediction model, and generate traffic flow prediction results for each grid. For details, refer to step S104 in the embodiment shown in Figure 1, and redundant explanations will be omitted here.

[0094] Step S805: A spatial overlay analysis is performed on the flood depth prediction result for each grid and the traffic flow prediction result for each grid, and urban traffic in the city grid is evaluated based on the spatial overlay analysis result, and an urban traffic evaluation result is generated.

[0095] Specifically, the above step S805 includes steps S8051 to S8053.

[0096] Step S8051: The flood depth prediction result for each grid is compared with a predetermined depth threshold to generate a flood depth level.

[0097] Specifically, in this embodiment, the unit of the grid crown water depth prediction result and the predetermined depth threshold can be uniformly set to millimeters. The predetermined depth threshold can be set to one or more. For example, the predetermined depth threshold is set to 0, 30, and 60. When the grid crown water depth prediction result H is H = 0, it indicates that the current urban grid has no crown water. When the grid crown water depth prediction result H is 0 < H < 30, it indicates that the current urban grid has shallow crown water. When the grid crown water depth prediction result H is 30 < H < 60, it indicates that the current urban grid has medium-scale crown water. When the grid crown water depth prediction result H is 60 < H, it indicates that the current urban grid has high crown water. Those skilled in the art can set the predetermined depth threshold and the crown water depth level according to the actual situation, and there is no limitation here.

[0098] Step S8052: Compare the traffic flow prediction result for each grid with a predetermined flow threshold to generate a traffic flow level.

[0099] Specifically, in this embodiment, the unit of the grid traffic flow prediction result and the predetermined flow threshold can be uniformly set to vehicles per hour. The predetermined flow threshold can be set to one or more. For example, the predetermined flow threshold is set to 0, 50, and 100. When the grid traffic flow prediction result Q is Q = 0, it indicates that the current urban grid has no traffic flow. When the grid traffic flow prediction result Q is 0 < Q < 50, it indicates that the current urban grid has low traffic flow. When the grid traffic flow prediction result Q is 50 < Q < 100, it indicates that the current urban grid has medium traffic flow. When the grid traffic flow prediction result Q is 100 < Q, it indicates that the current urban grid has high traffic flow. Those skilled in the art can set the predetermined flow threshold and the traffic flow level according to the actual situation, and there is no limitation here.

[0100] Step S8053: Determine the urban traffic danger level within the urban grid based on the crown water depth level and the traffic flow level, and use the urban traffic danger level as the urban traffic evaluation result.

[0101] Specifically, as shown in Table 2, the danger levels are divided into a total of six levels, ranging from no danger to very dangerous. Different danger levels can be reflected in different colors on each grid of the city, which can be displayed more intuitively by managers. The higher the danger level, the higher the priority of processing. Furthermore, the division of flow levels and flood levels can be specifically divided according to the actual conditions of different cities. In this embodiment, only the division of levels is shown here.

[0102] Table 2 JPEG2025026360000005.jpg71170

[0103] Note: "None" means no danger, "slightly low" means slightly low danger, "slightly dangerous" and "danger" have the same meaning, "slightly high" means slightly high danger, and "extreme" means extremely dangerous. There are a total of 6 danger levels, and the corresponding colors are danger warning colors. "(Hidden)" means that the area is free of danger, but if traffic flow increases suddenly or urban flooding increases, the danger level of the area will increase suddenly, and there is still hidden danger, so caution is required.

[0104] The urban traffic assessment method based on spatial overlay analysis in this embodiment obtains the danger levels of different areas of the city based on the flood depth level and the traffic flow level, and pre-arranges and directs the urban road traffic based on the predicted danger level, thereby maximally ensuring the property and personal safety of residents and preventing major damage to the development of the city.

[0105] As shown in FIG. 9, a specific embodiment of the urban traffic assessment method based on spatial overlay analysis is described below.

[0106] Example 1 In order to realize early warning and forecast of urban floods in advance from the hourly scale to the daily and weekly scales, and to predict the real-time flow rate of road traffic in advance, thereby arranging traffic transportation and arranging travel, the steps of the urban traffic assessment method based on spatial overlay analysis in this embodiment include the following steps:

[0107] (1) Data preparation and spatial element overlay analysis: In order to comprehensively predict urban floods, a research method of spatial element overlay analysis is proposed to provide a foundation for subsequent model construction and flood forecasting. This mainly involves data element preparation and urban grid division and spatial overlay of influencing elements.

[0108] <1> Preparing data elements.

[0109] The occurrence of urban flooding is related to many factors, and most of the previous studies on the occurrence of urban flooding focus on a single detailed analysis, but they ignore the potential contribution of other influencing factors to the occurrence of flooding. When considered from the spatial perspective, the influencing factors can be divided into three aspects: sky, surface, and underground.

[0110] 1) (Air) station monitoring rainfall data: It is a determining factor of flood disasters. Without precipitation, it is difficult for urban floods to occur (except coastal cities). Different precipitation distributions in urban precipitation (based on ground monitoring stations or radar rainfall monitoring) represent different spatial forms of precipitation (rainfall amount, rainfall duration, rainfall type, rainfall center and rainfall center movement trajectory) occurring above the city. Different spatial distributions of rainfall directly determine the occurrence of flood disasters in a city.

[0111] 2) (Surface) Urban surface topography and coverage data: In addition to the (air) station monitoring precipitation data, important factors to be considered are the different surface conditions of the urban surface (roads, forests, grasslands, residential areas) and the changing conditions of the land surface topography (ground gradient direction, topography). Traditional urban flood prediction focuses on precipitation factors, and many studies ignore the influence of the land surface factors of the city itself. Therefore, although the traditional research methods have a relatively high prediction accuracy of the possibility of flood occurrence based on precipitation factors, they ignore the influence of the land surface factors of the city, so it is generally difficult to accurately predict the specific areas where urban flooding will occur. Therefore, this embodiment also takes into account the spatial distribution conditions of the urban surface, providing a basis for accurate prediction of urban flood areas. The urban surface topography and coverage data includes elevation data, urban surface coverage data, urban topography, gradient data and urban road data.

[0112] 3) (Underground) Urban underground drainage pipeline data: including urban pipeline distribution data, urban pipeline caliber data, urban pipeline drainage data, urban drainage outlet distribution and sewage discharge capacity, etc. Urban underground drainage pipeline data can depict the flood response capacity of different areas of the city, and can also make detailed predictions about the possibility of flooding in different areas of the city.

[0113] Considering the spatial distribution of the main urban flood affecting factors in the sky, surface and underground of the city, urban flooding is analyzed by taking into account all spatial distribution scenes and performing overlay analysis, and the specific data is shown in Table 1.

[0114] <2> Splitting the city grid

[0115] The above describes the influencing factors on urban flood occurrence, where precipitation data is a time and space dual-varying factor, and urban surface factors and urban underground factors are spatially varying factors. For a certain precipitation, whether a city will have flood, the location of flood occurrence and the degree of flood damage depend on the spatial distribution of precipitation intensity, the spatial characteristic distribution of the urban surface, and the spatial distribution of the urban underground drainage characteristics. In order to take into account the spatial overlay of all influencing factors, in the form of grid division, grid division is performed with a predetermined size for the simulated urban boundary to obtain square grids (for example, 10m*10m), and the square grids are numbered. The numbered square grids are mapped to different areas of the city, thereby realizing spatial discretization of the city. The urban spatial discretization process provides a basis for subsequent regional prediction of flood risk in different areas of the city.

[0116] <3> Spatial overlay analysis of influencing factors

[0117] Dividing the prepared data into respective grid units according to spatial topological relationships; <2> Based on each square grid divided by , precipitation intensity data, precipitation duration data, urban surface land use type, urban slope, urban underground drainage pipeline width, drainage capacity, water retention capacity, etc. are obtained. <1> All of the information mentioned in is plotted on all grids of the city, all the influencing element attributes are assigned to the grids, and the spatial topology of all the elements that determine whether flooding will occur in the grids is overlaid on the grids, thereby realizing the spatial overlay processing of all the influencing elements in the space, and each unit grid represents the district position, district attributes and precipitation information of the city. Specifically, firstly, grid division is performed with a predetermined size for the simulated city boundary, then the grids are numbered and mapped to different districts of the city, and finally, all the prepared data is divided into each grid unit according to the spatial topology relationship, and each grid contains the spatial overlay information of the influencing elements in Table 1.

[0118] (2) Construction of urban flood model: Based on the above content of (1), that is, a spatial overlay data format for each grid of the city is constructed, and in order to predict the occurrence of urban flooding, an LSTM deep learning model is used to overlay spatial data of three aspects, namely, sky, surface, and underground, onto each grid unit of the city, thereby extracting scene data of urban flooding that has occurred in the unit in the past. One LSTM model is constructed for each constructed urban unit grid, and spatial trinity information for each constructed unit (sky, surface, and groundwater information, i.e., x0, x1, x2, ..., x in Figure 9) is extracted. t , x n ) to the LSTM deep learning model corresponding to each unit grid (i.e., A0, A1, A2, ..., A t , A n ) as input data, and the LSTM deep learning model corresponding to each unit grid is used to estimate the flood inundation depth for each city grid (i.e., h0, h1, h2, …, h t , h n ) and uses urban flood data (i.e., urban flood spatial distribution data and urban flood inundation depth monitoring data) as calibration data for each unit grid, which is used to train the LSTM model and obtain predictive simulation results.

[0119] By continuing to train the model, an LSTM urban flood inundation model can be obtained for each city grid, and the output of the model is the magnitude of the flood inundation depth H that will occur in that grid unit under precipitation input, where 0 indicates no inundation, and greater than 0 indicates that inundation depth will occur. Since each grid in the city outputs the magnitude of the flood inundation depth, the flood inundation distribution situation for the entire city can be obtained, and the construction of the urban flood inundation prediction model is completed.

[0120] (3) Construction of urban flood early warning and forecast model: based on the above content of (2), that is, the construction of a flood inundation model capable of predicting the spatial distribution of the city is completed. After the construction of the urban flood inundation model is completed, it can be applied to predict whether the city will have floods in the future and the area where they will occur. In order to realize this function, it is necessary to further construct and simulate a BP neural network model to predict future precipitation scenes. The input of the constructed LSTM model is the actual precipitation monitoring data of different stations distributed in the city. Since it is necessary to access meteorological forecast data for advance prediction, in order to estimate the future station monitoring data based on the future meteorological data, it is necessary to further construct a meteorological forecast model. It is necessary to establish a relationship between the report data and the precipitation spatial monitoring data, which is the second learning model constructed. Also, since the meteorological data is very related to the station monitoring data, the model construction process can be realized only by using an ordinary BP neural network, so that the future urban heavy rain flooding scenes can be predicted based on the weather forecast, and the predicted future urban flood occurrence scenes can correspond one-to-one to the hourly, daily, weekly and other scale data based on the weather forecast, that is, the urban flood forecast period always coincides with the meteorological forecast period. The grid-based BP neural network model is shown in Figure 7, where the input data is meteorological data (F) and the grid-based precipitation data is (P).

[0121] (4) Prediction of urban road traffic flow and urban traffic forecast

[0122] According to the predicted urban flood depth distribution, it is difficult to judge whether deeply flooded areas will have a significant impact on urban road traffic, because deeply flooded areas may have low traffic flow. Therefore, in order to predict the impact on urban road traffic, it is necessary to further obtain real-time forecast data of road traffic flow. By performing spatial overlay analysis on the predicted urban flood data and the forecast data of urban road traffic flow, it can be determined whether flooded areas will affect road traffic, and urban road traffic can be dispersed in a timely manner based on the predicted results, which mainly includes the following three aspects:

[0123] <1> Development of urban road traffic flow prediction model

[0124] The real-time road traffic flow of a city is related to its population distribution (residential population distribution), the time of day, the number of cars and light vehicles in the city, etc., but for a city, its road traffic flow is closely related to the date of the year and the time of the day. For example, in the winter of the year, the temperature is low, so the number of light vehicles decreases, and the number of residents who choose public transportation when going out increases. On the other hand, as the temperature rises, the corresponding number gradually increases. In addition, for example, on the day before the holidays such as May Day and National Day, many people travel by car, so the number of cars increases sharply throughout the day. In the day, 8-9 am is the morning rush hour, and 5-6 pm is the evening rush hour. The traffic flow in these two time periods increases sharply. According to the different population distribution laws of the city, the traffic flow in different areas of the corresponding city has a certain law. Considering all of this, the BP neural network model is adopted to predict the traffic flow in different areas of the city, as shown in Figure 2, (1) <2> Based on the grids divided by, the target city is divided into multiple city grids, and a BP neural network is constructed for each grid into which the city is divided. The city's date and time (month, day, hour) for the year (X1) and the number of residents in the grid (X2) are used as input information, and the traffic flow corresponding to the grid is used as output (Y) to train the BP neural network, thereby making it possible to obtain traffic flow prediction results at any time during the year. The model is trained based on traffic flow information from recent years (for example, the most recent five years).

[0125] <2> Spatial overlay analysis of urban road traffic flow and urban flooding prediction

[0126] <1> Based on the construction model of urban road traffic flow of different areas constructed by the above, in order to timely adjust urban traffic, a spatial overlay analysis is further carried out, that is, a spatial overlay analysis needs to be further carried out on the urban flood forecast inundation result and the predicted urban road traffic flow result. For the same urban area, when the flood depth forecast result is greater than a certain danger threshold and the predicted traffic flow of the area reaches a certain value, the urban area is a danger area. Based on the overlay relationship between the flood depth and the urban road traffic flow, the traffic danger of different areas can be warned. The results are shown in Table 2. The danger levels are divided into a total of six levels, from no danger to very dangerous. Different danger levels can be reflected in each grid of the city with different colors, which can be displayed intuitively by the manager. The higher the danger level, the higher the priority of processing. Moreover, the division of the flow level and the flood level can be specifically divided according to the actual situation of different cities.

[0127] <3> Timely arrangement of urban road traffic transportation based on the spatial overlay analysis results: Based on the above content of (3), the specific flood occurrence areas of the city in the future "hour-day-week" can be predicted. In this case, based on this prediction model, the areas that may cause serious danger can be predicted in a timely manner, the traffic within the "unit grid" can be dispersed, and the prediction result can be fed back to the traffic command department, so that the road section can be closed several hours in advance, and vehicles and personnel can be evacuated. However, since there is still a large uncertainty in the weather changes and predictions, when heavy rain is about to occur, the weather data can be interacted with in real time, and the model improvement results can be continuously corrected accordingly, and the traffic command strategy can be corrected in real time in conjunction with the traffic department to ensure the optimal road traffic conditions every time heavy rain occurs in the city, so as to avoid casualties and property losses of people caused by urban floods as much as possible, and realize the timely adjustment, optimization and layout of urban traffic. <2> Based on the danger levels of the spatial overlay analysis in the above, responses should be made in order from high to low. For extremely dangerous areas at the highest level, personnel in the current area should be notified to evacuate as soon as possible, traffic flow should be quickly relocated, and traffic control should be implemented to prevent personnel from entering, thereby downgrading the danger level. For non-dangerous areas at the lowest level, no management is required. However, in areas with high traffic flow and deep flooding, there is generally a possibility of hidden danger, so if there is still capacity in traffic command, management can be extended to non-key areas, so as to respond to emergencies in a more timely manner.

[0128] In this embodiment, based on the idea of ​​spatial overlay analysis and the idea of ​​urban spatial domain discretization, all factors affecting the flood inundation depth of the city, such as sky, surface, underground, etc., are overlaid on the same grid, and an LSTM model is built on each unit grid to realize the prediction of the flood inundation depth of different areas of the city. Then, in combination with the BP neural network model as a pre-processing step, the spatial distribution of urban flood in different time periods, such as hourly and daily, is predicted in the future in combination with weather forecasts. Finally, the BP neural network is built to predict the spatial distribution of flow rate of different time nodes in the city's traffic in the future, and the urban flood inundation depth and spatial overlay analysis are carried out to obtain the danger levels of different areas of the city, and the urban road traffic is pre-arranged and directed according to the predicted danger levels. Compared with the past, this method can support the transportation department in making more arrangements, thereby maximally ensuring the safety of residents' property and people, and preventing major damage to the development of the city.

[0129] In this embodiment, an urban traffic assessment device based on spatial overlay analysis is further provided, which is used to realize the above embodiment and preferred embodiment. For parts that have already been explained, duplicate explanations will be omitted. As used hereinafter, the term "module" can realize a combination of software and / or hardware of a certain function. The device described in the following embodiment is preferably realized in software, but it is also possible and envisioned to realize it in hardware or a combination of software and hardware.

[0130] This embodiment provides an urban traffic assessment device based on spatial overlay analysis, as shown in Figure 10: A first overlay module 1001 is used for obtaining station monitoring rainfall data, city surface data and city underground pipeline data in a target city area, and performing spatial data overlay on the station monitoring rainfall data, city surface data and city underground pipeline data in the target city area to generate spatial overlay data for each city grid; A first training module 1002 is used for training a predetermined deep learning model based on the spatial overlay data of each city grid to generate an urban flood inundation depth prediction model; A first acquisition module 1003 is used for acquiring weather forecast data in a target urban area, and determining a flood depth prediction result for each grid by using an urban flood inundation depth prediction model according to the weather forecast data; A first collection module 1004 is used for collecting urban population distribution data in a current period, inputting the urban population distribution data in the current period into an urban traffic flow prediction model, and generating a traffic flow prediction result for each grid; and a first evaluation module 1005, which is used to perform a spatial overlay analysis on the flood depth prediction result for each grid and the traffic flow prediction result for each grid, and evaluate urban traffic in the city grid based on the spatial overlay analysis result, and generate an urban traffic evaluation result.

[0131] In some alternative embodiments, the first overlay module 1001: a discretization processing sub-module, which is used for performing spatial discretization processing on the target urban area to generate a plurality of urban grids; an overlay sub-module, which is used to respectively divide the station monitoring rainfall data, the city surface data and the city underground pipeline data into multiple city grids according to the spatial topological relationship, and generate spatial overlay data for each city grid.

[0132] In some alternative embodiments, the first training module 1002 includes: a first input submodule, which is used to input the spatial overlay data of each city grid into a predetermined flood inundation deep learning model and generate an output result of the predetermined flood inundation deep learning model, and a predetermined flood inundation deep learning model is constructed for each city grid; and a first comparison submodule, which is used to acquire urban flood inundation distribution data, compare the output result of a predetermined flood inundation deep learning model with the urban flood inundation distribution data, and generate an urban flood inundation depth prediction model if the comparison result satisfies a predetermined condition.

[0133] In some alternative embodiments, the first acquisition module 1003: A second input sub-module, which is used for inputting the meteorological forecast dataset of the urban target area into the station precipitation forecast model and outputting station precipitation forecast data; and a third input sub-module, which is used for inputting the station precipitation forecast data into the urban flood inundation depth forecast model and outputting the inundation depth forecast result for each grid.

[0134] In some alternative embodiments, the urban flood inundation depth prediction model embodied by the first training module is a long-short-term memory deep learning model.

[0135] In some alternative embodiments, the first evaluation module 1005: A second comparison submodule is used for comparing the flood depth prediction result for each grid with a predetermined depth threshold to generate a flood depth level; a third comparison submodule, which is used for comparing the traffic flow prediction result for each grid with a predetermined traffic flow threshold value to generate a traffic flow level; a determining sub-module, which is used for determining an urban traffic danger level in the urban grid according to the flood depth level and the traffic flow level, and taking the urban traffic danger level as an urban traffic assessment result.

[0136] In some alternative embodiments, The system further includes a dispatch module, which is used for dispatching urban road traffic in real time and providing early risk warning based on the spatial overlay analysis result.

[0137] Further functional descriptions of the above modules and units are the same as those of the corresponding embodiments, and therefore repeated descriptions will be omitted here.

[0138] The urban traffic assessment apparatus based on spatial overlay analysis in this embodiment is presented in the form of a functional unit, where unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory executing one or more software or fixed programs, and / or other device capable of providing the above-mentioned functionality.

[0139] An embodiment of the present invention further provides a computer device, which comprises an urban traffic assessment apparatus based on spatial overlay analysis as shown in FIG.

[0140] As shown in FIG. 11, FIG. 11 is a structural schematic diagram of a computer device according to an alternative embodiment of the present invention. As shown in FIG. 11, the computer device includes one or more processors 10, a memory 20, and an interface for connecting each component, including a high-speed interface and a low-speed interface, and each component is communicatively connected to each other via different buses, and may be attached to a common motherboard or attached in other ways as needed. The processor can process instructions executed within the computer device, and includes a memory or instructions stored on the memory for displaying graphical information of a GUI on an external input / output device (e.g., a display device coupled to the interface). In some alternative embodiments, multiple processors and / or multiple buses can be used with multiple memories and multiple memories as needed. Similarly, multiple computer devices may be connected, each device providing partially required operations (e.g., as a server array, a set of blade servers, or a multiprocessor system). Take one processor 10 as an example in FIG. 11.

[0141] The processor 10 may be a central processing unit, a network processor, or a combination thereof. The processor 10 may further include a hardware chip. The hardware chip may be an application specific integrated circuit, a programmable logic device, or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable logic gate array, a general purpose array logic, or any combination thereof.

[0142] The memory 20 stores instructions executable by the at least one processor 10, thereby causing the at least one processor 10 to execute and realize the methods shown in the above embodiments.

[0143] The memory 20 may include a program storage area for storing an operating system, an application program required for at least one function, and a data storage area for storing data generated in response to use of the computer device. The memory 20 may also include high-speed random access memory, and may further include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some alternative embodiments, the memory 20 may include memory that is selectively located remotely with respect to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of such networks include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0144] Memory 20 may include volatile memory, such as random access memory, memory may include non-volatile memory, such as flash memory, hardware or solid state drives, or memory 20 may include a combination of the above types of memory.

[0145] The computer device further includes an input device 30 and an output device 40. The processor 10, the memory 20, the input device 30 and the output device 40 may be connected by a bus or in other ways, and in FIG. 11, they are connected by a bus as an example.

[0146] The input device 30 can receive input numeric or character information and generate key signal inputs for user settings and function control of the computing device, such as a touch screen, a keypad, a mouse, a track pad, a touch pad, an indicator stick, one or more mouse buttons, a track ball, a joystick, etc. The output device 40 may include a display device, an auxiliary lighting device (e.g., LED), and a tactile feedback device (e.g., vibration motor), etc. The display device includes, but is not limited to, a liquid crystal display, a light emitting diode display, and a plasma display. In some alternative embodiments, the display device may be a touch screen.

[0147] The embodiments of the present invention further provide a computer readable storage medium, and the methods according to the embodiments of the present invention may be implemented in hardware, firmware, recordable in a storage medium, or as computer code downloaded over a network, originally stored in a remote storage medium or a non-transitory machine readable storage medium, but stored in a local storage medium, whereby the methods described herein may be processed by software stored in the storage medium using a general purpose computer, a dedicated processor, or programmable or dedicated hardware. The storage medium may be a magnetic disk, an optical disk, a read only memory, a random access memory, a flash memory memory, a hardware or solid state drive, etc., and further, the storage medium may further include a combination of the above types of memory. It is understood that the computer, processor, microprocessor controller, or programmable hardware may include a storage component capable of storing or receiving the software or computer code, and when the software or computer code is accessed and executed by the computer, processor, or hardware, it realizes the methods shown in the embodiments.

[0148] Although the embodiments of the present invention have been described with reference to the drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present invention, and all such modifications and variations fall within the scope defined by the appended claims.

Claims

1. A method for urban transportation assessment based on spatial overlay analysis, comprising: Obtaining station monitoring rainfall data, city surface data and city underground pipeline data in a target urban area, and performing spatial data overlay on the station monitoring rainfall data, the city surface data and the city underground pipeline data in the target urban area to generate spatial overlay data for each urban grid; training a deep learning model based on the spatial overlay data for each urban grid to generate an urban flood inundation depth prediction model; obtaining weather forecast data within a target urban area, and determining a flood depth prediction result for each grid using the urban flood inundation depth prediction model based on the weather forecast data; collecting urban population distribution data for a current period, inputting the urban population distribution data for the current period into an urban traffic flow forecasting model, and generating a traffic flow forecasting result for each grid; performing a spatial overlay analysis on the flood depth prediction result for each grid and the traffic flow prediction result for each grid, and evaluating urban traffic in the city grid based on the spatial overlay analysis result, and generating an urban traffic evaluation result; The step of performing spatial data overlay on the station monitoring rainfall data, the city surface data, and the city underground pipeline data within a target urban area to generate spatial overlay data for each urban grid includes: performing a spatial discretization process on the target urban area to generate a plurality of urban grids; respectively dividing the station monitoring rainfall data, the city surface data and the city underground pipeline data into the multiple city grids according to spatial topological relationships, and generating spatial overlay data for each city grid; based on each divided city grid, plotting all the information of precipitation intensity data, precipitation duration data, city surface land use type, city gradient, city underground drainage pipeline width, drainage capacity and water retention capacity into all grids of the city, and giving all the influencing element attributes to the grids, thereby realizing the spatial overlay processing of all the influencing elements in space; The step of determining a flood depth prediction result for each grid using the urban flood inundation depth prediction model based on the meteorological forecast data includes: inputting the weather forecast dataset of the urban target area into a station precipitation forecast model and outputting station precipitation forecast data; inputting the station precipitation forecast data into the urban flood inundation depth forecast model and outputting a flood depth forecast result for each grid; The step of training a deep learning model based on the spatial overlay data for each urban grid to generate an urban flood inundation depth prediction model includes: inputting the spatial overlay data for each city grid into a predetermined flood inundation deep learning model to generate an output result of the predetermined flood inundation deep learning model, where one predetermined flood inundation deep learning model is constructed for each city grid; acquiring urban flood inundation distribution data, comparing the output results of the predetermined flood inundation deep learning model with the urban flood inundation distribution data, and generating the urban flood inundation depth prediction model if the comparison result satisfies a predetermined condition.

2. The step of determining a flood depth prediction result for each grid using the urban flood inundation depth prediction model based on the meteorological forecast data includes: inputting the weather forecast dataset of the urban target area into a station precipitation forecast model and outputting station precipitation forecast data; 2. The method of claim 1, further comprising: inputting the station precipitation forecast data into the urban flood inundation depth forecast model and outputting a flood depth forecast result for each grid.

3. The method of claim 1, wherein the urban flood inundation depth prediction model is a long-short-term memory deep learning model.

4. The step of performing a spatial overlay analysis on the flood depth prediction result for each grid and the traffic flow prediction result for each grid, evaluating urban traffic in the city grid based on the spatial overlay analysis result, and generating an urban traffic evaluation result, comparing the flood depth prediction result for each grid with a predetermined depth threshold to generate a flood depth level; comparing the traffic flow prediction result for each grid with a predetermined flow threshold to generate a traffic flow level; 2. The method according to claim 1, further comprising: determining an urban traffic danger level in a city grid based on the flood depth level and the traffic flow level; and taking the urban traffic danger level as an urban traffic assessment result.

5. The method according to claim 1 , further comprising: real-time urban road traffic dispatch and hazard early warning according to the spatial overlay analysis result.

6. An urban traffic assessment device based on spatial overlay analysis, comprising: A first overlay module is used for obtaining station monitoring rainfall data, city surface data and city underground pipeline data in a target city area, and performing spatial data overlay on the station monitoring rainfall data, the city surface data and the city underground pipeline data in the target city area, to generate spatial overlay data for each city grid; A first training module is used for training a predetermined deep learning model based on the spatial overlay data for each city grid to generate an urban flood inundation depth prediction model; A first acquisition module is used for acquiring weather forecast data within a target urban area, and determining a flood depth prediction result for each grid using the urban flood inundation depth prediction model according to the weather forecast data; A first collection module is used for collecting urban population distribution data in a current period, inputting the urban population distribution data in the current period into an urban traffic flow prediction model, and generating a traffic flow prediction result for each grid; a first evaluation module, which is used for performing a spatial overlay analysis on the flood depth prediction result for each grid and the traffic flow prediction result for each grid, and evaluating urban traffic in the city grid based on the spatial overlay analysis result, and generating an urban traffic evaluation result; The first overlay module includes: a discretization processing submodule, which is used for performing spatial discretization processing on the target urban area to generate a plurality of urban grids; an overlay sub-module for dividing the station monitoring rainfall data, the city surface data and the city underground pipeline data into the multiple city grids according to spatial topology relationships, respectively, and generating spatial overlay data for each city grid; the overlay sub-module for plotting all of the information of precipitation intensity data, precipitation duration data, city surface land use type, city gradient, city underground drainage pipeline width, drainage capacity and water retention capacity into all grids of the city according to each divided city grid, and assigning all of the influencing element attributes to the grids, thereby realizing the spatial overlay processing of all of the influencing elements in space; The first training module includes: a first input submodule for inputting the spatial overlay data for each city grid into a predetermined flood inundation deep learning model to generate an output result of the predetermined flood inundation deep learning model, in which one predetermined flood inundation deep learning model is constructed for each city grid; a first comparison submodule, which is used to acquire urban flood inundation distribution data, compare the output result of the predetermined flood inundation deep learning model with the urban flood inundation distribution data, and generate the urban flood inundation depth prediction model if the comparison result satisfies a predetermined condition.

7. The first acquisition module is A second input sub-module, which is used for inputting the meteorological forecast dataset of the urban target area into the station precipitation forecast model, and outputting station precipitation forecast data; The apparatus of claim 6, further comprising: a third input sub-module, which is used for inputting the station precipitation forecast data into the urban flood inundation depth forecast model and outputting the inundation depth forecast result for each grid.

8. The urban flood inundation depth prediction model specifically included in the first training module is a long-short-term memory deep learning model.

9. The first evaluation module: A second comparison submodule is used for comparing the flood depth prediction result for each grid with a predetermined depth threshold to generate a flood depth level; A third comparison submodule is used for comparing the traffic flow prediction result for each grid with a predetermined traffic flow threshold value to generate a traffic flow level; The apparatus of claim 6, further comprising: a determining sub-module, which is used for determining an urban traffic danger level in a city grid based on the flood depth level and the traffic flow level, and taking the urban traffic danger level as an urban traffic assessment result.

10. The apparatus as claimed in claim 6, further comprising a dispatch module used for real-time dispatch of urban road traffic according to the spatial overlay analysis result and providing hazard early warning.

11. 1. A computer device comprising: A computer device comprising: a memory; and a processor, the memory and the processor being communicatively connected to each other; the memory storing computer instructions; and the processor executing the computer instructions to perform the urban transportation evaluation method based on spatial overlay analysis according to any one of claims 1 to 5.

12. A computer-readable storage medium having stored thereon computer instructions for causing a computer to execute the urban traffic assessment method based on spatial overlay analysis according to any one of claims 1 to 5.

Images