Flood prevention early warning system based on artificial intelligence

Abstract

This invention discloses an artificial intelligence-based flood warning system, relating to the field of flood monitoring technology. It includes: a database for storing basic structural parameters at various locations of a target dam and constructing a dam structural model based on these parameters; a data acquisition module deployed at various locations of the target dam to acquire corresponding hydrological parameters and import them into the dam structural model; a data processing module for processing the acquired hydrological parameters to obtain hydrological time-series data; a data analysis module for analyzing the hydrological time-series data and determining the presence of flood risk based on the analysis results; and a flood warning module for generating corresponding flood warning information when a flood risk exists. By combining the dam structural model with shear coefficients and consolidation coefficients, a flood control assessment value is obtained, thereby more realistically reflecting the actual stress and consolidation changes caused by hydraulic impacts such as water pressure and flow velocity on different sections of the dam.

Description

Technical Field

[0001] This invention relates to the field of flood control monitoring technology, specifically to a flood control early warning system based on artificial intelligence. Background Technology

[0002] Dikes are the core infrastructure of the flood control and disaster reduction system. With the frequent occurrence of extreme weather and the violent fluctuations in water levels during the flood season, higher requirements are placed on the safety monitoring and early warning of dikes. Most existing flood warning systems focus on the collection of macro-hydrological data and simple threshold alarms, such as relying solely on the overall water level of the river to trigger an early warning. Existing technologies have the following obvious shortcomings: 1. They fail to combine dynamic hydrological data with the physical structural characteristics of the dam itself, thus failing to accurately reflect the actual stress and consolidation changes caused by hydraulic impacts such as water pressure and flow velocity on different sections of the dam; 2. They lack refined spatial analysis, making it difficult to capture the risk of local scour damage caused by sudden changes in local flow velocity or special angles of flow, and failing to accurately indicate the specific location and spread of the danger. Therefore, an artificial intelligence-based flood warning system is now provided. Summary of the Invention

[0003] The purpose of this invention is to provide a flood prevention and early warning system based on artificial intelligence.

[0004] The objective of this invention can be achieved through the following technical solution: an artificial intelligence-based flood warning system, comprising: The database is used to store the basic structural parameters of each location of the target dam and to build a dam structural model based on the basic structural parameters; The data acquisition module is deployed at various locations on the target dam to obtain the corresponding hydrological parameters and import the obtained hydrological parameters into the dam structure model. The data processing module is used to process the obtained hydrological parameters to obtain hydrological time series data; The data analysis module is used to analyze hydrological time-series data and determine whether there is a risk of flooding based on the analysis results; The flood warning module is used to generate corresponding flood warning information when there is a risk of flooding.

[0005] Furthermore, the basic structural parameters include: dam crest width, dam crest elevation, upstream slope ratio, downstream slope ratio, porosity, and density; A physical model of the target dam is constructed based on the structural parameters, including dam crest width, dam crest elevation, upstream slope ratio, and downstream slope ratio. Within the physical model of the target dam, the water-facing slope surface of the target dam is divided into several sub-surface regions; The porosity and density of each sub-surface region are obtained, and the porosity and density are correlated with the corresponding sub-surface region. The correlation results are then mapped into the physical model of the target dam to obtain the dam structure model of the target dam.

[0006] Furthermore, the data acquisition module consists of several data acquisition terminals; At the center of each sub-surface area corresponding to the target dam, a corresponding data acquisition terminal is deployed, and an associated virtual data acquisition node is generated based on the corresponding position of the deployed data acquisition terminal within the dam structure model. Hydrological parameters at the corresponding location are acquired in real time through the deployed data acquisition terminals; The hydrological parameters include water pressure, flow velocity, water depth, and water flow direction; The hydrological parameters of each sub-surface region are imported into the corresponding virtual data acquisition nodes within the dam structure model.

[0007] Furthermore, the data processing module processes the obtained hydrological parameters to obtain hydrological time-series data. The process includes: Construct time axes corresponding to each sub-surface region, and generate corresponding water pressure change curves, flow velocity change curves, and water depth change curves based on the obtained hydrological parameters of each sub-surface region. The corresponding angle of attack variation curve is generated based on the angle between the water flow direction and the sub-surface area; The generated water pressure change curve, flow velocity change curve, water depth change curve, and flow angle change curve are mapped onto the time axis; Set a time window and mark the various change curves within the time window as the corresponding hydrological time series data for the corresponding sub-surface area.

[0008] Furthermore, the data analysis module analyzes hydrological time-series data, and the process of determining whether there is a risk of flooding based on the analysis results includes: The water pressure change curve, flow velocity change curve, and water depth change curve within the time window are traversed separately, and the maximum and minimum water pressure values ​​in the water pressure change curve, the maximum and minimum flow velocity values ​​in the flow velocity change curve, and the maximum and minimum water depth values ​​in the water depth change curve are obtained respectively. Further obtain the time corresponding to the maximum water pressure value, minimum water pressure value, maximum flow velocity value, minimum flow velocity value, maximum water depth value, and minimum water depth value; Then the shear resistance coefficient of each sub-surface region within the time window is obtained; Furthermore, based on the porosity and density corresponding to the sub-surface region, the consolidation coefficient of the sub-surface region is obtained; Based on the obtained shear strength and consolidation coefficient, the flood resistance assessment value of this sub-surface region is obtained, denoted as . ; Set a flood control threshold, denoted as K0; The obtained flood control assessment value is compared with the set flood control threshold, and the flood risk of the sub-surface area is determined based on the comparison results. like If K < K0, it means the corresponding sub-surface region is normal; otherwise, it means the corresponding sub-surface region is abnormal. The analysis results are then used to generate corresponding evaluation labels at the corresponding locations within the dam structure model. These evaluation labels include anomaly labels and normal labels. Anomaly labels correspond to anomalies in the sub-surface area, while normal labels correspond to normal sub-surface areas.

[0009] Furthermore, the process by which the flood warning module generates corresponding flood warning information when there is a risk of flooding includes: Obtain the anomaly labels in the dam structure model and determine the anomaly distribution areas; Obtain the regional span value of the abnormal distribution area; The risk assessment value of the abnormal distribution area is obtained based on the area span value of the abnormal distribution area; Several risk warning threshold ranges are set, and each risk threshold warning range corresponds to a warning level signal; The obtained risk assessment values ​​are matched with the risk warning threshold range, and the warning level of the abnormal distribution area is generated based on the matching results. The warning levels and corresponding abnormal distribution areas are summarized to generate corresponding flood warning information.

[0010] Furthermore, the process of determining the abnormal distribution area includes: Select any sub-surface region corresponding to an anomaly label as the reference region; Obtain the adjacent regions of the baseline region. If there are abnormal labels in the adjacent regions, merge the adjacent regions with abnormal labels with the baseline region to form a new baseline region. Then, check whether there are abnormal labels in the adjacent regions of the new baseline region. Continue in this way until there are no abnormal labels in the adjacent regions of the baseline region, and obtain the abnormal distribution region.

[0011] Furthermore, the regional span value includes a longitudinal span value and a transverse span value, wherein the longitudinal span value refers to the maximum elevation difference of the abnormal distribution area, and the transverse span value refers to the maximum horizontal difference.

[0012] Compared with the prior art, the beneficial effects of the present invention are: 1. By constructing a dam structure model that integrates physical parameters and hydrological time series parameters, and combining shear strength coefficient and consolidation coefficient to calculate flood control assessment value, dynamic hydrological time series parameters are combined with the dam structure model to more realistically reflect the actual stress and consolidation changes of water pressure, flow velocity and other hydraulic impacts on different sections of the dam. 2. By identifying “abnormal distribution areas” and combining the longitudinal and lateral span values ​​of the areas for secondary risk quantification, differentiated early warning levels are finally output, thereby obtaining the location, spatial scale and severity of possible flood situations. Attached Figure Description

[0013] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0014] Figure 1 This is a schematic diagram of the present invention. Detailed Implementation

[0015] like Figure 1 As shown, the artificial intelligence-based flood warning system includes: The database is used to store the basic structural parameters of each location of the target dam and to build a dam structural model based on the basic structural parameters; The data acquisition module is deployed at various locations on the target dam to obtain the corresponding hydrological parameters and import the obtained hydrological parameters into the dam structure model. The data processing module is used to process the obtained hydrological parameters to obtain hydrological time series data; The data analysis module is used to analyze hydrological time-series data and determine whether there is a risk of flooding based on the analysis results; The flood warning module is used to generate corresponding flood warning information when there is a risk of flooding.

[0016] It should be further explained that, in the specific implementation process, the basic structural parameters include: dam crest width, dam crest elevation, upstream slope ratio, downstream slope ratio, porosity, and density; A physical model of the target dam is constructed based on the structural parameters, including dam crest width, dam crest elevation, upstream slope ratio, and downstream slope ratio. Within the physical model of the target dam, the water-facing slope surface of the target dam is divided into several sub-surface regions, and each sub-surface region is rectangular with the same area. The porosity and density of each sub-surface region are obtained, and the porosity and density are correlated with the corresponding sub-surface region. The correlation results are then mapped into the physical model of the target dam to obtain the dam structure model of the target dam.

[0017] It should be further explained that, in the specific implementation process, the data acquisition module obtains hydrological parameters through the following steps: The data acquisition module consists of several data acquisition terminals; At the center of each sub-surface area corresponding to the target dam, a corresponding data acquisition terminal is deployed, and an associated virtual data acquisition node is generated based on the corresponding position of the deployed data acquisition terminal within the dam structure model. Hydrological parameters at the corresponding location are acquired in real time through the deployed data acquisition terminals; The hydrological parameters include water pressure, flow velocity, water depth, and flow direction. It should be noted that the hydrological parameters obtained by each data acquisition terminal are the hydrological parameters of the location where the data acquisition terminal is located. For example, water pressure refers to the water pressure detected at the location of the data acquisition terminal, and water depth refers to the water depth at the location of the data acquisition terminal. The hydrological parameters of each sub-surface region are imported into the corresponding virtual data acquisition nodes within the dam structure model.

[0018] It should be further explained that, in the specific implementation process, the data processing module processes the obtained hydrological parameters to obtain hydrological time-series data. The process includes: Construct time axes corresponding to each sub-surface region, and generate corresponding water pressure change curves, flow velocity change curves, and water depth change curves based on the obtained hydrological parameters of each sub-surface region. The corresponding angle of attack variation curve is generated based on the angle between the water flow direction and the sub-surface area; The generated water pressure change curve, flow velocity change curve, water depth change curve, and flow angle change curve are mapped onto the time axis; Set a time window, denoted as [t1, t2], where t2 corresponds to the current time and t1 is the previous time. Each change curve within the time window is marked and used as the corresponding hydrological time series data for the corresponding sub-surface region.

[0019] It should be further explained that, in the specific implementation process, the data analysis module analyzes hydrological time-series data, and the process of determining whether there is a risk of flooding based on the analysis results includes: The water pressure change curve, flow velocity change curve, and water depth change curve within the time window are traversed separately, and the maximum and minimum water pressure values ​​in the water pressure change curve, the maximum and minimum flow velocity values ​​in the flow velocity change curve, and the maximum and minimum water depth values ​​in the water depth change curve are obtained respectively. Each sub-surface region is labeled as i, where i = 1, 2, ..., n; The maximum water pressure value of the sub-surface region labeled i is then denoted as... The minimum water pressure value is denoted as The maximum flow velocity value is denoted as The minimum flow velocity is denoted as The maximum water depth is denoted as The minimum water depth is denoted as ; Then, the times corresponding to the maximum water pressure, minimum water pressure, maximum flow velocity, minimum flow velocity, maximum water depth, and minimum water depth are further obtained and denoted as follows: , , , , , ; Then the shear resistance coefficient of the sub-surface region labeled i within the time window is obtained, denoted as . ,in: ; in This is the pressure conversion coefficient, and the pressure conversion coefficient is related to the maximum angle of attack within the time window; Furthermore, the porosity and density corresponding to the sub-surface region labeled i are denoted as follows: and ; Based on the obtained porosity and density, the consolidation coefficient of this sub-surface region is obtained, denoted as . ,in: ; Where Yp is the slope ratio of the upstream slope of the target dam; Based on the obtained shear strength and consolidation coefficient, the flood resistance assessment value of this sub-surface region is obtained, denoted as . ,in: ; Set a flood control threshold, denoted as K0; The obtained flood control assessment value is compared with the set flood control threshold, and the flood risk of the sub-surface area is determined based on the comparison results. like If K < K0, it means the corresponding sub-surface region is normal; otherwise, it means the corresponding sub-surface region is abnormal. The analysis results are then used to generate corresponding evaluation labels at the corresponding locations within the dam structure model. These evaluation labels include anomaly labels and normal labels. Anomaly labels correspond to anomalies in the sub-surface area, while normal labels correspond to normal sub-surface areas.

[0020] It should be further explained that, in the specific implementation process, the process by which the flood warning module generates corresponding flood warning information when there is a risk of flooding includes: Obtain the anomaly labels in the dam structure model, and select the sub-surface region corresponding to any anomaly label as the reference region; Obtain the neighboring regions of the baseline region. If there are abnormal labels in the neighboring regions, merge the neighboring regions with abnormal labels with the baseline region to form a new baseline region. Then, check whether there are abnormal labels in the neighboring regions of the new baseline region. Continue in this way until there are no abnormal labels in the neighboring regions of the baseline region, and obtain the abnormal distribution region. Obtain the regional span value of the abnormal distribution area, which includes the longitudinal span value and the lateral span value, wherein the longitudinal span value refers to the maximum elevation difference of the abnormal distribution area, and the lateral span value refers to the maximum horizontal difference. The risk assessment value of the abnormal distribution area is then obtained, denoted as Fp, where: ; in, This represents the horizontal span value. h1 is the longitudinal span value, h2 is the distance between the horizontal line where the transverse span value is located and the bottom of the dam, a is the minimum value between the abnormal distribution area and the bottom of the dam, M is the natural constant, and M is the number of sub-surface areas involved in the abnormal distribution area. Several risk warning threshold ranges are set, and each risk threshold warning range corresponds to a warning level signal; The obtained risk assessment values ​​are matched with the risk warning threshold range, and the warning level of the abnormal distribution area is generated based on the matching results. The warning levels and corresponding abnormal distribution areas are summarized to generate corresponding flood warning information.

[0021] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any modifications or equivalent substitutions made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A flood prevention and early warning system based on artificial intelligence, characterized in that, include: The database is used to store the basic structural parameters of each location of the target dam and to build a dam structural model based on the basic structural parameters; The data acquisition module is deployed at various locations on the target dam to obtain the corresponding hydrological parameters and import the obtained hydrological parameters into the dam structure model. The data processing module is used to process the obtained hydrological parameters to obtain hydrological time series data; The data analysis module is used to analyze hydrological time-series data and determine whether there is a risk of flooding based on the analysis results; The flood warning module is used to generate corresponding flood warning information when there is a risk of flooding.

2. The flood early warning system based on artificial intelligence according to claim 1, characterized in that, The basic structural parameters include: dam crest width, dam crest elevation, upstream slope ratio, downstream slope ratio, porosity, and density; A physical model of the target dam is constructed based on the structural parameters, including dam crest width, dam crest elevation, upstream slope ratio, and downstream slope ratio. Within the physical model of the target dam, the water-facing slope surface of the target dam is divided into several sub-surface regions; The porosity and density of each sub-surface region are obtained, and the porosity and density are correlated with the corresponding sub-surface region. The correlation results are then mapped into the physical model of the target dam to obtain the dam structure model of the target dam.

3. The flood early warning system based on artificial intelligence according to claim 2, characterized in that, The data acquisition module consists of several data acquisition terminals; At the center of each sub-surface area corresponding to the target dam, a corresponding data acquisition terminal is deployed, and an associated virtual data acquisition node is generated based on the corresponding position of the deployed data acquisition terminal within the dam structure model. Hydrological parameters at the corresponding location are acquired in real time through the deployed data acquisition terminals; The hydrological parameters include water pressure, flow velocity, water depth, and water flow direction; The hydrological parameters of each sub-surface region are imported into the corresponding virtual data acquisition node within the dam structure model.

4. The artificial intelligence-based flood early warning system according to claim 3, characterized in that, The data processing module processes the obtained hydrological parameters to obtain hydrological time-series data. The process includes: Construct time axes corresponding to each sub-surface region, and generate corresponding water pressure change curves, flow velocity change curves, and water depth change curves based on the obtained hydrological parameters of each sub-surface region. The corresponding angle of attack variation curve is generated based on the angle between the water flow direction and the sub-surface area; The generated water pressure change curve, flow velocity change curve, water depth change curve, and flow angle change curve are mapped onto the time axis; Set a time window and mark the various change curves within the time window as the corresponding hydrological time series data for the corresponding sub-surface area.

5. The flood early warning system based on artificial intelligence according to claim 4, characterized in that, The data analysis module analyzes hydrological time-series data and determines whether there is a risk of flooding based on the analysis results. This process includes: The water pressure change curve, flow velocity change curve, and water depth change curve within the time window are traversed separately, and the maximum and minimum water pressure values ​​in the water pressure change curve, the maximum and minimum flow velocity values ​​in the flow velocity change curve, and the maximum and minimum water depth values ​​in the water depth change curve are obtained respectively. Further obtain the time corresponding to the maximum water pressure value, minimum water pressure value, maximum flow velocity value, minimum flow velocity value, maximum water depth value, and minimum water depth value; Then the shear resistance coefficient of each sub-surface region within the time window is obtained; Furthermore, based on the porosity and density corresponding to the sub-surface region, the consolidation coefficient of the sub-surface region is obtained; Based on the obtained shear strength and consolidation coefficient, the flood resistance assessment value of this sub-surface region is obtained, denoted as . ; Set a flood control threshold, denoted as K0; The obtained flood control assessment value is compared with the set flood control threshold, and the flood risk of the sub-surface area is determined based on the comparison results. like If K < K0, it means the corresponding sub-surface region is normal; otherwise, it means the corresponding sub-surface region is abnormal. The analysis results are then used to generate corresponding evaluation labels at the corresponding locations within the dam structure model. These evaluation labels include anomaly labels and normal labels. Anomaly labels correspond to anomalies in the sub-surface area, while normal labels correspond to normal sub-surface areas.

6. The flood early warning system based on artificial intelligence according to claim 5, characterized in that, The process by which the flood warning module generates corresponding flood warning information when there is a risk of flooding includes: Obtain the anomaly labels in the dam structure model and determine the anomaly distribution areas; Obtain the regional span value of the abnormal distribution area; The risk assessment value of the abnormal distribution area is obtained based on the area span value of the abnormal distribution area; Several risk warning threshold ranges are set, and each risk threshold warning range corresponds to a warning level signal; The obtained risk assessment values ​​are matched with the risk warning threshold range, and the warning level of the abnormal distribution area is generated based on the matching results. The warning levels and corresponding abnormal distribution areas are summarized to generate corresponding flood warning information.

7. The flood early warning system based on artificial intelligence according to claim 6, characterized in that, The process of determining the abnormal distribution area includes: Select any sub-surface region corresponding to an anomaly label as the reference region; Obtain the adjacent regions of the baseline region. If there are abnormal labels in the adjacent regions, merge the adjacent regions with abnormal labels with the baseline region to form a new baseline region. Then, check whether there are abnormal labels in the adjacent regions of the new baseline region. Continue in this way until there are no abnormal labels in the adjacent regions of the baseline region, and obtain the abnormal distribution region.

8. The artificial intelligence-based flood early warning system according to claim 6, characterized in that, The area span value includes a longitudinal span value and a transverse span value, wherein the longitudinal span value refers to the maximum elevation difference of the abnormal distribution area, and the transverse span value refers to the maximum horizontal difference.

Images