An urban flood risk monitoring system and method

Abstract

The present invention discloses an urban flood risk monitoring system and method. The monitoring system includes a rainfall statistics module for obtaining historical rainfall data of the city; a data acquisition module for obtaining urban waterlogging data; a data analysis module for judging the flood occurrence probability according to the historical rainfall data of the city and the urban waterlogging data, and generating a warning instruction to be transmitted to the flood warning module; and a flood warning module for performing different forms of warnings according to the warning instruction. The present invention realizes flood risk monitoring by statistically analyzing the historical rainfall data of the city and collecting the waterlogging data of the current rainfall, constructing a flood prediction model using a BP neural network model, determining the rainfall type according to the water level rising rate, and obtaining the flood occurrence probability according to the water level value corresponding to the flood occurrence of the rainfall type. At the same time, different forms of warnings are given for different levels of flood risks, effectively avoiding the occurrence of flood disasters.

Description

Technical Field

[0001] The present invention belongs to the field of flood prevention, and particularly relates to an urban flood risk monitoring system and method. Background Art

[0002] Flood is a natural phenomenon with high peak and large volume, and rapid rise of water level. Waterlogging is a natural phenomenon due to long-term precipitation or heavy rain that cannot be drained into river channels and ditches in time, forming surface water accumulation. When floods and waterlogging cause losses to humans, they become flood disasters.

[0003] In recent years, with the development of social economy, the main part of flood disaster losses has shifted to cities, and the characteristics of floods have also changed greatly. Many cities are along rivers, lakes, coasts or near mountains and waters. Some cities are located in low-lying plains and are often threatened by floods. Compared with rural areas, the population and assets in cities are highly concentrated, and the disaster losses are much greater. Therefore, there is a need for urban flood monitoring.

[0004] In traditional urban flood monitoring, water level data is mostly obtained through sensors, and whether there is a flood risk is directly judged based on the water level data. This method has a single data source and cannot accurately obtain the rainfall type and the probability of flood occurrence, resulting in inaccurate judgment results being common. Summary of the Invention

[0005] The purpose of the present invention is to provide an urban flood risk monitoring system and method to solve the problems existing in the above-mentioned prior art.

[0006] On the one hand, to achieve the above purpose, the present invention provides an urban flood risk monitoring system, including a rainfall statistics module, a data acquisition module, a data analysis module, and a flood warning module;

[0007] The rainfall statistics module is used to obtain historical rainfall data of the city;

[0008] The data acquisition module is used to obtain urban water accumulation data;

[0009] The data analysis module is used to judge the probability of flood occurrence according to the historical rainfall data of the city and the urban water accumulation data, and generate a warning instruction to transmit to the flood warning module;

[0010] The flood warning module is used to give different forms of warnings according to the warning instruction.

[0011] Optionally, the urban historical rainfall data includes rainfall type, rainfall characteristics, flood water level value, rainfall duration, and water level rising rate; the rainfall type is the main type of urban rainfall, the rainfall characteristics are the rainfall intensity corresponding to the rainfall type; the water level rising rate is the ratio of the flood water level value to the rainfall duration; the urban waterlogging data includes the waterlogging water level value and the waterlogging water level rising rate.

[0012] Optionally, the data analysis module constructs a flood prediction model using a BP neural network, trains the model with the historical rainfall data and the urban waterlogging data as inputs, and outputs the flood occurrence probability.

[0013] Optionally, the flood prediction model determines the rainfall type according to the waterlogging water level rising rate in the urban waterlogging data, the water level rising rate in the rainfall data, and the rainfall intensity, presets an expected probability value according to the rainfall type, randomly assigns weights to each connection layer, sets a flood threshold according to the flood water level value corresponding to the rainfall type, obtains the first flood occurrence probability according to the ratio of the waterlogging water level value in the urban waterlogging data to the flood threshold, compares the output flood occurrence probability with the preset expected probability value to obtain the model error, updates the randomly assigned weights according to the model error, and repeats the calculation to reduce the error. When the model converges, the flood occurrence probability is output through the output layer.

[0014] Optionally, when the flood occurrence probability reaches 70%, the data analysis module generates a flood prevention instruction and transmits it to the flood warning module. When the flood occurrence probability reaches 90%, the data analysis module generates an emergency warning instruction and transmits it to the flood warning module.

[0015] Optionally, the flood warning module includes a warning light and an audio device. When a flood prevention instruction is obtained, the warning light is controlled to flash; when an emergency warning instruction is obtained, the audio device is controlled to give an audio alarm.

[0016] On the other hand, to achieve the above object, the present invention provides a method for monitoring urban flood risk, including the following steps:

[0017] Statistical urban historical rainfall data, the urban historical rainfall data includes rainfall type, rainfall characteristics, flood water level value, rainfall duration, and water level rising rate; the rainfall type is the main type of urban rainfall, the rainfall characteristics are the rainfall intensity corresponding to the rainfall type; the water level rising rate is the ratio of the flood water level value to the rainfall duration;

[0018] Collect urban waterlogging data, the urban waterlogging data includes the waterlogging water level value and the waterlogging water level rising rate;

[0019] Build a flood prediction model based on the BP neural network. Based on the flood prediction model, use the historical rainfall data and the urban waterlogging data as inputs for training to obtain the flood occurrence probability;

[0020] Carry out different forms of early warnings based on the flood occurrence probability.

[0021] Optionally, the process of using the historical rainfall data and the urban waterlogging data as inputs for training includes:

[0022] Determine the rainfall type based on the rising rate of the waterlogging level in the urban waterlogging data and the rising rate of the water level in the rainfall data, as well as the rainfall intensity;

[0023] Preset an expected probability value based on the rainfall type, randomly assign weights to each connection layer, and set a flood threshold based on the flood water level value corresponding to the rainfall type;

[0024] Obtain the first flood occurrence probability based on the ratio of the waterlogging level value in the urban waterlogging data to the flood threshold, compare the first flood occurrence probability with the preset expected probability value to obtain the model error;

[0025] Update the randomly assigned weights based on the model error, repeat the calculation to reduce the error, and when the model converges, output the flood occurrence probability through the output layer.

[0026] Optionally, when the flood occurrence probability reaches 70%, carry out a warning in the form of lights; when the flood occurrence probability reaches 90%, carry out an audio prompt warning.

[0027] The technical effect of the present invention is:

[0028] The present invention proposes a system and method for monitoring urban flood risks. By statistically analyzing the historical rainfall data of the city and collecting the waterlogging data of the current rainfall, a flood prediction model is constructed using a BP neural network model. The rainfall type is determined based on the rising rate of the waterlogging, and the flood occurrence probability is obtained according to the water level value corresponding to the rainfall type when a flood occurs, realizing flood risk monitoring. At the same time, by carrying out different forms of early warnings for different levels of flood risks, the occurrence of flood disasters can be effectively avoided. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] The drawings constituting a part of this application are used to provide a further understanding of this application. The schematic embodiments of this application and their descriptions are used to explain this application and do not constitute an improper limitation of this application. In the drawings:

[0030] Figure 1 It is a schematic structural diagram of the urban flood risk monitoring system in the embodiment of the present invention;

[0031] Figure 2 This is the flowchart of the urban flood risk monitoring method in the embodiments of the present invention. Specific implementation manners

[0032] It should be noted that, without conflict, the embodiments in this application and the features in the embodiments can be combined with each other. The following will describe this application in detail with reference to the drawings and in combination with the embodiments.

[0033] It should be noted that the steps shown in the flowchart of the drawings can be executed in a computer system such as a set of computer-executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in a different order than here.

[0034] Embodiment 1

[0035] As Figure 1 shown, this embodiment provides an urban flood risk monitoring system, including a rainfall statistics module, a data acquisition module, a data analysis module, and a flood warning module; specifically:

[0036] The rainfall statistics module is used to obtain the historical rainfall data of the city; the historical rainfall data of the city refers to the main rainfall types, rainfall characteristics, water level values at the time of flood, rainfall duration, and the rising rate of the water level in the city when floods occur, where the rainfall characteristic is the rainfall intensity corresponding to the rainfall type, and the rising rate of the water level is calculated by the rainfall statistics module through the ratio of the obtained flood water level value to the rainfall duration.

[0037] The data acquisition module is used to obtain the water accumulation data at the moment of rainfall in the city; the water accumulation data is the water level value and the rising rate of the water accumulation in the city during this rainfall process.

[0038] After obtaining the historical rainfall data of the city and the waterlogging data of the current rainfall, the data analysis module constructs a flood prediction model by using a BP neural network, and uses the historical rainfall data and the waterlogging data of the city as inputs for model training, that is, outputs the flood occurrence probability; the BP neural network is one of the most widely used neural network models at present, including an input layer, a hidden layer and an output layer. The specific process of the flood prediction model obtaining the flood occurrence probability includes: first, input the waterlogging data and the historical rainfall data of the city into the input layer of the neural network. In the hidden layer, determine the rainfall type according to the rising rate of the waterlogging level in the waterlogging data and the rainfall intensity and the rising rate in the rainfall data, preset the expected probability value according to the rainfall type, randomly assign weights to each connection layer, and set a flood threshold according to the flood water level value corresponding to the rainfall type. Calculate according to the ratio of the waterlogging level value in the waterlogging data of the city to the flood threshold to obtain the first flood occurrence probability. Compare the first flood occurrence probability with the preset expected probability value to obtain the model error, update the randomly assigned weights according to the model error, and repeat the calculation to reduce the error. When the model converges, the accurate flood occurrence probability is output through the output layer.

[0039] As a preferred embodiment of the present application, when it is determined that the flood occurrence probability is 70%, the data analysis module generates a flood prevention instruction and transmits it to the flood warning module. When the flood occurrence probability is 90%, the data analysis module generates an emergency warning instruction and transmits it to the flood warning module. The specific threshold settings in this paragraph are artificially set according to actual needs, and all belong to the protection scope of the present application.

[0040] The flood warning module includes a plurality of warning lights and a portable audio device. When obtaining a flood prevention instruction, the flood warning module controls the warning lights to flash for warning, notifying the staff to take preventive measures in advance; when obtaining an emergency warning instruction, the flood warning module controls the portable audio device to emit sound for audio alarm, notifying the staff carrying the audio device to take preventive measures immediately.

[0041] Embodiment 2

[0042] As Figure 2 shown, a method for monitoring urban flood risk is provided in this embodiment, including the following steps:

[0043] Statistical historical rainfall data of the city, the historical rainfall data of the city including rainfall type, rainfall characteristics, flood water level value, rainfall duration, and rising rate of water level; the rainfall type is the main type of urban rainfall, the rainfall characteristics are the rainfall intensity corresponding to the rainfall type; the rising rate of water level is the ratio of the flood water level value to the rainfall duration;

[0044] Collect urban waterlogging data, where the urban waterlogging data includes waterlogging level values and the rising rate of the waterlogging level;

[0045] Construct a flood prediction model based on a BP neural network. Based on the flood prediction model, use the historical rainfall data and the urban waterlogging data as inputs for training to obtain the probability of flood occurrence;

[0046] Give different forms of warnings based on the probability of flood occurrence.

[0047] As a preferred embodiment of the present application, the process of using the historical rainfall data and the urban waterlogging data as inputs for training includes:

[0048] Determine the rainfall type based on the rising rate of the waterlogging level in the urban waterlogging data, the rising rate of the water level in the rainfall data, and the rainfall intensity;

[0049] Preset an expected probability value based on the rainfall type, randomly assign weights to each connection layer, and set a flood threshold based on the flood level value corresponding to the rainfall type;

[0050] Obtain the first flood occurrence probability based on the ratio of the waterlogging level value in the urban waterlogging data to the flood threshold, compare the first flood occurrence probability with the preset expected probability value to obtain the model error;

[0051] Update the randomly assigned weights based on the model error, repeat the calculation to reduce the error, and when the model converges, output the flood occurrence probability through the output layer.

[0052] As a preferred embodiment of the present application, when the flood occurrence probability reaches 70%, give a warning in the form of lights; when the flood occurrence probability reaches 90%, give an audio prompt warning.

[0053] The various embodiments in this specification are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. The same or similar parts among the various embodiments can be referred to each other. For the methods disclosed in the embodiments, since they correspond to the systems disclosed in the embodiments, the description is relatively simple. For the relevant parts, refer to the description of the system part.

[0054] The above is only a preferred specific embodiment of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by those skilled in the art within the technical scope disclosed in the present application should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. An urban flood risk monitoring system, characterized in that, It includes a rainfall statistics module, a data collection module, a data analysis module, and a flood warning module; The rainfall statistics module is used to obtain the historical rainfall data of the city; The data collection module is used to obtain the urban waterlogging data; The data analysis module is used to judge the flood occurrence probability according to the historical rainfall data of the city and the urban waterlogging data, and generate a warning instruction to transmit to the flood warning module; The data analysis module uses a BP neural network to construct a flood prediction model, takes the historical rainfall data and the urban waterlogging data as inputs for model training, and outputs the flood occurrence probability; The flood prediction model determines the rainfall type according to the rising rate of the waterlogging level in the urban waterlogging data and the rising rate of the water level in the rainfall data, as well as the rainfall intensity, presets an expected probability value according to the rainfall type, randomly assigns weights to each connection layer, sets a flood threshold according to the flood water level value corresponding to the rainfall type, obtains the first flood occurrence probability according to the ratio of the waterlogging level value in the urban waterlogging data to the flood threshold, compares the output flood occurrence probability with the preset expected probability value to obtain the model error, updates the randomly assigned weights according to the model error, and repeats the calculation to reduce the error. When the model converges, the flood occurrence probability is output through the output layer; The flood warning module is used to issue warnings in different forms according to the warning instruction.

2. The urban flood risk monitoring system according to claim 1, wherein The historical rainfall data of the city includes rainfall type, rainfall characteristics, flood water level value, rainfall duration, and water level rising rate; the rainfall type is the main type of urban rainfall, the rainfall characteristics are the rainfall intensity corresponding to the rainfall type; the water level rising rate is the ratio of the flood water level value to the rainfall duration; the urban waterlogging data includes the waterlogging level value and the rising rate of the waterlogging level.

3. The urban flood risk monitoring system according to claim 1, wherein When the flood occurrence probability reaches 70%, the data analysis module generates a flood prevention instruction to transmit to the flood warning module. When the flood occurrence probability reaches 90%, the data analysis module generates an emergency warning instruction to transmit to the flood warning module.

4. The urban flood risk monitoring system according to claim 1, characterized in that, The flood warning module includes a warning light and an audio device. When a flood prevention instruction is obtained, the warning light is controlled to flash; when an emergency warning instruction is obtained, the audio device is controlled to give an audio alarm.

5. A method for monitoring urban flood risk, characterized in that, It includes the following steps: Statistical analysis of the historical rainfall data of the city. The historical rainfall data of the city includes rainfall type, rainfall characteristics, flood water level value, rainfall duration, and water level rising rate; the rainfall type is the main type of urban rainfall, the rainfall characteristics are the rainfall intensity corresponding to the rainfall type; the water level rising rate is the ratio of the flood water level value to the rainfall duration; Collect urban waterlogging data. The urban waterlogging data includes the waterlogging level value and the rising rate of the waterlogging level; Construct a flood prediction model based on a BP neural network. Based on the flood prediction model, use the historical rainfall data and the urban waterlogging data as inputs for training to obtain the flood occurrence probability; The process of using the historical rainfall data and the urban waterlogging data as inputs for training includes: Determine the rainfall type based on the rising rate of the waterlogging level in the urban waterlogging data and the rising rate of the water level in the rainfall data, as well as the rainfall intensity; Preset an expected probability value based on the rainfall type, randomly assign weights to each connection layer, and set a flood threshold based on the flood water level value corresponding to the rainfall type; Obtain the first flood occurrence probability based on the ratio of the waterlogging level value in the urban waterlogging data to the flood threshold, and compare the first flood occurrence probability with the preset expected probability value to obtain the model error; Update the randomly assigned weights based on the model error, repeat the calculation to reduce the error, and when the model converges, output the flood occurrence probability through the output layer; Conduct different forms of early warnings based on the flood occurrence probability; When the flood occurrence probability reaches 70%, conduct a warning in the form of lights; when the flood occurrence probability reaches 90%, conduct an audio prompt warning.

Images