CN114565603A - Integrated flood disaster accurate monitoring and early warning method - Google Patents
Integrated flood disaster accurate monitoring and early warning method Download PDFInfo
- Publication number
- CN114565603A CN114565603A CN202210230779.0A CN202210230779A CN114565603A CN 114565603 A CN114565603 A CN 114565603A CN 202210230779 A CN202210230779 A CN 202210230779A CN 114565603 A CN114565603 A CN 114565603A
- Authority
- CN
- China
- Prior art keywords
- prediction
- information
- data
- network model
- model
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Withdrawn
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
- G06T17/05—Geographic models
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10032—Satellite or aerial image; Remote sensing
- G06T2207/10044—Radar image
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A50/00—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE in human health protection, e.g. against extreme weather
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Software Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Molecular Biology (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Geometry (AREA)
- Quality & Reliability (AREA)
- Remote Sensing (AREA)
- Computer Graphics (AREA)
- Alarm Systems (AREA)
Abstract
The invention discloses an integrated flood disaster accurate monitoring and early warning method, which belongs to the field of disaster monitoring and comprises the following specific steps: (1) collecting information of each area; (2) flood disaster information prediction; (3) performing statistical planning on each region; (4) analyzing and optimizing the information prediction; according to the invention, the water level information is monitored in real time by arranging the laser radar, so that the estimation accuracy of the water level can be greatly improved, the collection can be kept stable under different illumination conditions, the prediction neural model can be automatically updated, the manual update of workers is not needed, the workload of the workers is reduced, and the working efficiency of the workers is improved.
Description
Technical Field
The invention relates to the field of disaster monitoring, in particular to an integrated accurate flood disaster monitoring and early warning method.
Background
Flood disasters comprise flood disasters and waterlogging disasters, wherein the disasters caused by the increase of water quantity in rivers, lakes and coastal waters, the flooding caused by the rise of water level and the outburst of mountain floods are called flood disasters due to strong rainfall, ice and snow melting, ice slush, dam burst and storm surge; the flood disasters such as lands, houses and the like caused by waterlogging and flooding are often caused by the fact that a large amount of accumulated water and runoff are generated due to too concentrated heavy rain, heavy rain or long-term rainfall and drainage is not timely, so that the disasters such as the flood disasters and the rain disasters are often simultaneously or continuously generated in the same area and sometimes difficult to accurately define, and are often collectively called as the flood disasters, and the accidents with serious disaster loss are frequently rare due to the lack of thought and material preparation for the occurrence of the flood disasters. Therefore, the flood disaster risk monitoring and early warning evaluation is enhanced, and the disaster risk management is carried out scientifically, so that the method has very important practical significance;
through retrieval, the Chinese patent No. CN113240688A discloses an integrated flood disaster accurate monitoring and early warning method, which can effectively overcome the defects that comprehensive and integrated heterogeneous multi-scale monitoring data cannot be effectively utilized, and the problems of spatial non-uniformity, dynamic risk change and multi-level risk of the flood disaster cannot be effectively solved, but the estimation accuracy of the water level cannot be guaranteed in an image detection mode, and the phenomenon of instability is caused under different illumination conditions; in addition, the existing integrated flood disaster accurate monitoring and early warning method needs workers to manually update the prediction model, so that the workload of the workers is increased, and the working efficiency of the workers is reduced; therefore, an integrated flood disaster accurate monitoring and early warning method is provided.
Disclosure of Invention
The invention aims to solve the defects in the prior art and provides an integrated flood disaster accurate monitoring and early warning method.
In order to achieve the purpose, the invention adopts the following technical scheme:
an integrated flood disaster accurate monitoring and early warning method comprises the following specific steps:
(1) collecting information of each area: the computing platform performs region segmentation on the regions, performs data capture on basic environment information of each region after segmentation, and performs information annotation on captured environment data:
(2) flood disaster information prediction: constructing a prediction network model, carrying out model training on the prediction network model, importing environment information of each area into the trained prediction network model for disaster prediction, and uploading a prediction result to a service terminal for storage;
(3) and (3) performing statistical planning on each area: the computing platform counts flood control equipment and resident personnel in each region according to the prediction result, and meanwhile, distributes and plans statistical data;
(4) analyzing and optimizing information prediction: and carrying out flood information collection on the area at regular intervals, checking and analyzing the prediction result, and carrying out training optimization processing on the prediction network model according to the analysis result.
As a further aspect of the present invention, the region segmentation in step (1) specifically comprises the following steps:
the method comprises the following steps: the computing platform is in communication connection with the remote sensing satellite and receives remote sensing images of all areas transmitted by the remote sensing satellite in real time;
step two: and carrying out image optimization processing on the remote sensing image, analyzing and checking the landform of each region, and segmenting the remote sensing image of the region according to different landforms.
As a further scheme of the present invention, the terrain of the area in the second step mainly includes plains, plateaus, hills, mountains and basins.
As a further scheme of the invention, the information annotation in the step (1) comprises the following specific steps:
the first step is as follows: the computing platform constructs a relevant three-dimensional model according to the collected environmental information, and marks the landform of each area on the three-dimensional model;
the second step is that: collecting water level information in real time through a laser radar water level gauge, meanwhile, collecting precipitation in the area periodically, and extracting precipitation and water level information in the past area from a main server;
the third step: and marking the latest water level information, the regional rainfall information and the rising or falling information of the two groups of information on the three-dimensional model, and updating each group of information in real time.
As a further scheme of the invention, the model training in the step (2) comprises the following specific steps:
s1.1: collecting N groups of observation data, selecting one observation data from the N groups of observation data as verification data, fitting a test model by using the rest observation data, verifying the precision of the model by using the observation value which is excluded firstly through root mean square error, and repeating the steps for N times;
s1.2: the method comprises the steps that a computing platform extracts past regional flood data in a main server, establishes a data sample, selects any subset as a test set for each group of data, then takes the rest subsets as a training set, counts the root mean square error again until each group of data is subjected to primary prediction, and outputs the data with the best prediction result as optimal parameters;
s1.3: and (3) carrying out standardization processing on the training data set according to the optimal parameters, finally, conveying the training samples to a learning network model, automatically setting specific parameters of the model, and training the model by adopting a long-term iteration method.
As a further scheme of the present invention, the allocation plan in step (3) specifically includes the following steps:
s2.1: registering the flood control equipment which is counted in each area, and analyzing the quality of each flood control equipment;
s2.2: reporting the flood control equipment with higher damage rate to related departments, and simultaneously prompting related workers to replace the damaged flood control equipment;
s2.3: the computing platform estimates the flood disaster range, counts residents in the estimated range and plans a safety region;
s2.4: and the computing platform sends out flood early warning to residents in the estimated range and simultaneously feeds back the position of each safety region to each resident.
As a further scheme of the invention, the training optimization in the step (4) comprises the following specific steps:
q1: drawing a corresponding judgment curve according to the prediction accuracy of the prediction network model, analyzing the corresponding judgment curve, and extracting information of various factors of prediction errors;
q2: leading error factors into an input layer of a prediction network model, leading accurate prediction data into the input layer of the prediction network model, and establishing an RBF neural network model;
q3: the RBF neural network model compares error factors with accurate prediction data and eliminates redundant data in the error factors;
q4: integrating the two models of the prediction network model and the RBF neural network model, constructing an integrated model, optimizing the prediction rule through the integrated model according to processed error factors, and finally importing the optimal prediction rule as an output result into the prediction network model for rule replacement.
Compared with the prior art, the invention has the beneficial effects that:
1. compared with the traditional early warning method, the integrated flood disaster accurate monitoring and early warning method has the advantages that remote sensing images of all regions are divided according to different terrains through the computing platform, relevant three-dimensional models are built according to collected environment information, meanwhile, the computing platform collects water level information in real time through the laser radar water level gauge, meanwhile, precipitation of the regions is collected regularly, precipitation and water level information of the previous regions are extracted from the main server, meanwhile, latest water level information, region precipitation information and rising or falling information of two groups of information are marked on the three-dimensional models, meanwhile, all groups of information are updated in real time, the water level information is monitored in real time through the arrangement of the laser radar, the estimation accuracy of the water level can be greatly improved, and meanwhile, the collection stability can be kept under different illumination conditions;
2. the integrated flood disaster accurate monitoring and early warning method draws a corresponding judgment curve according to the prediction accuracy of the prediction network model, extracts information of various factors of prediction errors, introduces error factors into an input layer of the prediction network model, introduces accurate prediction data into the input layer of the prediction network model, establishes an RBF neural network model, eliminates redundant data in the error factors, integrates the prediction network model and the RBF neural network model, constructs an integrated model, optimizes the prediction rules according to the processed error factors through the integrated model, and finally introduces the optimal prediction rules into the prediction network model as output results for rule replacement, so that the prediction neural model can be automatically updated without manual update of workers, and the workload of the workers is reduced, the working efficiency of workers is improved.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the principles of the invention and not to limit the invention.
Fig. 1 is a flow chart of an integrated flood disaster accurate monitoring and early warning method provided by the invention.
Detailed Description
Example 1
Referring to fig. 1, an integrated flood disaster accurate monitoring and early warning method specifically discloses a disaster prediction method:
collecting information of each area: and the computing platform performs region segmentation on the regions, performs data capture on the basic environment information of each segmented region, and performs information annotation on the captured environment data.
Specifically, the computing platform is in communication connection with the remote sensing satellite and receives remote sensing images of various areas transmitted by the remote sensing satellite in real time, after receiving is completed, the computing platform performs image optimization processing on the remote sensing images, analyzes and checks landforms of various areas, and segments the remote sensing images of the areas according to different landforms.
Specifically, firstly, a relevant three-dimensional model is built by a computing platform according to collected environment information, all regional terrains are marked on the three-dimensional model, after marking is completed, the computing platform collects water level information in real time through a laser radar water level gauge, precipitation in the region is collected regularly, past regional precipitation and water level information are extracted from a main server, meanwhile, latest water level information, regional precipitation information and two groups of information rising or falling information are marked on the three-dimensional model, all groups of information are updated in real time, the water level information is monitored in real time through the arrangement of the laser radar, the estimation accuracy of the water level can be greatly improved, and meanwhile, collection stability can be kept under different illumination conditions.
It should be further noted that the terrain of the region mainly includes plains, plateaus, hills, mountains and basins.
Flood disaster information prediction: and constructing a prediction network model, carrying out model training on the prediction network model, importing the environment information of each area into the trained prediction network model for disaster prediction, and uploading the prediction result to a service terminal for storage.
Specifically, a computing platform collects N groups of observation data, selects one observation data from the N groups of observation data as verification data, uses the rest observation data to fit a test model, uses the observation value which is excluded firstly to verify the precision of the model through root mean square error, repeats the operation for N times, extracts the past regional flooding data in a main server and establishes a data sample after the repetition, selects any subset as a test set for each group of data, then selects the rest subsets as a training set, counts the root mean square error again until each group of data is predicted once, outputs the data with the best prediction result as the optimal parameters, standardizes the training data set according to the optimal parameters, finally transmits the training sample to a learning network model, and sets the specific parameters of the model by itself, and the model is trained by adopting a long-term iteration method.
Example 2
Referring to fig. 1, an integrated flood disaster accurate monitoring and early warning method specifically discloses a training optimization method, except for the same structure as the above embodiment:
and (3) performing statistical planning on each area: and the computing platform counts the flood control equipment and the resident in each region according to the prediction result, and simultaneously performs distribution planning on the statistical data.
Specifically, the computing platform registers flood control equipment which is counted and completed in each area, analyzes the quality of each flood control equipment, reports flood control equipment with high damage rate to relevant departments, prompts relevant workers to replace damaged flood control equipment, estimates flood disaster ranges after the flood control equipment is counted and counts residents in the estimated ranges, plans the safe areas, sends flood early warning to residents in the estimated ranges, and feeds back the positions of the safe areas to the residents.
Analyzing and optimizing information prediction: and (4) periodically collecting flood information of the area, checking and analyzing the prediction result, and training and optimizing the prediction network model according to the analysis result.
Specifically, the calculation platform draws a corresponding judgment curve according to the prediction accuracy of the prediction network model, analyzes the curve, extracts various factors of prediction errors, introduces error factors into an input layer of the prediction network model, introduces accurate prediction data into the input layer of the prediction network model, establishes an RBF neural network model, compares the error factors with the accurate prediction data by the RBF neural network model after the construction is completed, eliminates redundant data in the error factors, integrates the prediction network model and the RBF neural network model, constructs an integrated model, optimizes the prediction rules according to the processed error factors by the integrated model, introduces the optimal prediction rules into the prediction network model as an output result for rule replacement, and can update the prediction neural model by itself, need not staff's manual update, reduce staff's work load, improve staff work efficiency.
Claims (7)
1. An integrated flood disaster accurate monitoring and early warning method is characterized by comprising the following specific steps:
(1) collecting information of each area: the computing platform performs region segmentation on the regions, performs data capture on basic environment information of each region after segmentation, and performs information annotation on captured environment data:
(2) flood disaster information prediction: constructing a prediction network model, carrying out model training on the prediction network model, importing environment information of each area into the trained prediction network model for disaster prediction, and uploading a prediction result to a service terminal for storage;
(3) and (3) performing statistical planning on each area: the computing platform counts flood control equipment and resident personnel in each region according to the prediction result, and meanwhile, distributes and plans statistical data;
(4) analyzing and optimizing information prediction: and (4) periodically collecting flood information of the area, checking and analyzing the prediction result, and training and optimizing the prediction network model according to the analysis result.
2. The integrated flood disaster accurate monitoring and early warning method according to claim 1, wherein the region segmentation in the step (1) comprises the following specific steps:
the method comprises the following steps: the computing platform is in communication connection with the remote sensing satellite and receives remote sensing images of all areas transmitted by the remote sensing satellite in real time;
step two: and carrying out image optimization processing on the remote sensing image, analyzing and checking the landform of each area, and segmenting the remote sensing image of the area according to different landforms.
3. The integrated flood disaster accurate monitoring and early warning method according to claim 2, wherein the regional landforms in the second step mainly comprise plains, plateaus, hills, mountains and basins.
4. The integrated flood disaster accurate monitoring and early warning method according to claim 1, wherein the information labeling in the step (1) comprises the following specific steps:
the first step is as follows: the computing platform constructs a relevant three-dimensional model according to the collected environmental information, and marks the landform of each area on the three-dimensional model;
the second step is that: collecting water level information in real time through a laser radar water level gauge, meanwhile, collecting precipitation in the area periodically, and extracting precipitation and water level information in the past area from a main server;
the third step: and marking the latest water level information, the regional rainfall information and the rising or falling information of the two groups of information on the three-dimensional model, and updating each group of information in real time.
5. The integrated flood disaster accurate monitoring and early warning method according to claim 1, wherein the model training in the step (2) specifically comprises the following steps:
s1.1: collecting N groups of observation data, selecting one observation data from the N groups of observation data as verification data, fitting a test model by using the rest observation data, verifying the precision of the model by using the observation value which is excluded firstly through root mean square error, and repeating the steps for N times;
s1.2: the method comprises the steps that a computing platform extracts past regional flood data in a main server, data samples are established, any subset is selected as a test set for each group of data, then remaining subsets are taken as training sets, root mean square errors are counted again until each group of data is predicted once, and the data with the best prediction results are output as optimal parameters;
s1.3: and (3) carrying out standardization processing on the training data set according to the optimal parameters, finally, conveying the training samples to a learning network model, automatically setting specific parameters of the model, and training the model by adopting a long-term iteration method.
6. The integrated flood disaster accurate monitoring and early warning method according to claim 1, wherein the distribution planning in the step (3) comprises the following specific steps:
s2.1: registering the flood control equipment which is counted in each area, and analyzing the quality of each flood control equipment;
s2.2: reporting the flood control equipment with higher damage rate to related departments, and simultaneously prompting related workers to replace the damaged flood control equipment;
s2.3: the computing platform estimates the flood disaster range, counts residents in the estimated range and plans a safety region;
s2.4: and the computing platform sends out flood early warning to residents in the estimated range, and simultaneously feeds back the position of each safety region to each resident.
7. The integrated flood disaster accurate monitoring and early warning method according to claim 1, wherein the training optimization in the step (4) specifically comprises the following steps:
q1: drawing a corresponding judgment curve according to the prediction accuracy of the prediction network model, analyzing the curve and extracting information of various factors of prediction errors;
q2: leading error factors into an input layer of a prediction network model, leading accurate prediction data into the input layer of the prediction network model, and establishing an RBF neural network model;
q3: the RBF neural network model compares error factors with accurate prediction data and eliminates redundant data in the error factors;
q4: integrating the two models of the prediction network model and the RBF neural network model, constructing an integrated model, optimizing the prediction rule through the integrated model according to processed error factors, and finally importing the optimal prediction rule as an output result into the prediction network model for rule replacement.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210230779.0A CN114565603A (en) | 2022-03-10 | 2022-03-10 | Integrated flood disaster accurate monitoring and early warning method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210230779.0A CN114565603A (en) | 2022-03-10 | 2022-03-10 | Integrated flood disaster accurate monitoring and early warning method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN114565603A true CN114565603A (en) | 2022-05-31 |
Family
ID=81718352
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210230779.0A Withdrawn CN114565603A (en) | 2022-03-10 | 2022-03-10 | Integrated flood disaster accurate monitoring and early warning method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114565603A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115329670A (en) * | 2022-08-11 | 2022-11-11 | 深圳朗道智通科技有限公司 | Data acquisition method for unmanned vehicle |
CN117315510A (en) * | 2023-09-25 | 2023-12-29 | 广东省核工业地质局测绘院 | Ecological environment mapping system based on remote sensing interpretation |
-
2022
- 2022-03-10 CN CN202210230779.0A patent/CN114565603A/en not_active Withdrawn
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115329670A (en) * | 2022-08-11 | 2022-11-11 | 深圳朗道智通科技有限公司 | Data acquisition method for unmanned vehicle |
CN117315510A (en) * | 2023-09-25 | 2023-12-29 | 广东省核工业地质局测绘院 | Ecological environment mapping system based on remote sensing interpretation |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109508630B (en) | Method for identifying water level of water gauge based on artificial intelligence | |
CN111507375B (en) | Urban waterlogging risk rapid assessment method and system | |
CN114565603A (en) | Integrated flood disaster accurate monitoring and early warning method | |
CN113919231B (en) | PM2.5 concentration space-time change prediction method and system based on space-time diagram neural network | |
CN110646867A (en) | Urban drainage monitoring and early warning method and system | |
CN106382471A (en) | Municipal drainage pipe network diagnostic assessment method giving consideration to key node | |
CN112016831A (en) | AI intelligent forecast-based urban waterlogging landing area identification method | |
CN111866452B (en) | Video real-time monitoring method for emergency disposal of landslide and damming dam | |
CN113392523B (en) | Sewage pipe network health condition diagnosis method based on long-duration multi-measuring-point | |
CN117010726B (en) | Intelligent early warning method and system for urban flood control | |
CN112528563B (en) | Urban waterlogging early warning method based on SVM algorithm | |
CN109186706A (en) | A method of for the early warning of Urban Storm Flood flooding area | |
CN110134907B (en) | Rainfall missing data filling method and system and electronic equipment | |
CN111928120A (en) | Rain and sewage pipe network monitoring system | |
CN110119790A (en) | The method of shared bicycle quantity statistics and density estimation based on computer vision | |
CN113128709A (en) | Water supply pipe network leakage management system | |
CN114021766A (en) | Urban water safety prediction method based on Internet of things and big data | |
CN115204688A (en) | Comprehensive evaluation method for health of drainage system | |
CN116227362A (en) | Municipal drainage pipe network state prediction method based on graph and deep learning | |
CN115909662A (en) | Check dam flood control early warning method and system | |
CN115685389A (en) | Lower concave bridge accumulated water prediction method based on downscaling and dynamic division of catchment area | |
CN114118870A (en) | Downstream flood estimation method for coupling iced lake burst flood and snowmelt flood | |
CN116502805B (en) | Scheduling scheme rapid screening method based on surrounding area water network lifting quantitative evaluation model | |
CN116737989B (en) | Urban surface hydrologic monitoring network construction method based on video image | |
CN116126963A (en) | Inflow infiltration diagnosis method for nonlinear multi-source heterogeneous drainage model |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
WW01 | Invention patent application withdrawn after publication | ||
WW01 | Invention patent application withdrawn after publication |
Application publication date: 20220531 |