CN112308292B - Method for drawing fire risk grade distribution map - Google Patents
Method for drawing fire risk grade distribution map Download PDFInfo
- Publication number
- CN112308292B CN112308292B CN202011077194.7A CN202011077194A CN112308292B CN 112308292 B CN112308292 B CN 112308292B CN 202011077194 A CN202011077194 A CN 202011077194A CN 112308292 B CN112308292 B CN 112308292B
- Authority
- CN
- China
- Prior art keywords
- fire
- data
- grid
- risk
- city
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/29—Geographical information databases
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- 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
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0635—Risk analysis of enterprise or organisation activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T11/00—2D [Two Dimensional] image generation
- G06T11/20—Drawing from basic elements, e.g. lines or circles
- G06T11/206—Drawing of charts or graphs
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Evolutionary Computation (AREA)
- Entrepreneurship & Innovation (AREA)
- Artificial Intelligence (AREA)
- Operations Research (AREA)
- Software Systems (AREA)
- Tourism & Hospitality (AREA)
- Quality & Reliability (AREA)
- Databases & Information Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Marketing (AREA)
- Game Theory and Decision Science (AREA)
- General Business, Economics & Management (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Development Economics (AREA)
- Computer Hardware Design (AREA)
- General Health & Medical Sciences (AREA)
- Educational Administration (AREA)
- Geometry (AREA)
- Health & Medical Sciences (AREA)
- Remote Sensing (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Medical Informatics (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
Abstract
The invention discloses a method for drawing a fire risk level distribution map, which comprises the following steps: selecting a city for drawing a fire grade distribution map, and collecting fire influence factor data and fire condition data of a historical fire point place in the city to establish a fire information distribution condition data set; marking and extracting various landforms in the region in the city to obtain grid data; preprocessing a fire information distribution condition data set by using a data cleaning and feature extraction method; training the preprocessed fire information distribution condition data set by adopting a machine learning algorithm to obtain a plurality of prediction models; calculating the grid data by using the prediction models to obtain a prediction value corresponding to each grid; determining the weight of each predicted value, calculating a final risk value, and performing normalization processing and grade division; and drawing a fire risk grade distribution map based on the grid data and the divided grades.
Description
Technical Field
The invention relates to the field of fire prevention and control management, in particular to a method for drawing a fire risk level distribution diagram.
Background
With the rapid development of economy and the rapid advance of urbanization process, the fire conditions of modern society are increasingly complex, and the relative lag of infrastructure construction, frequent fire accidents and continuously increased fire risks bring more and more serious influence on the sustainable development of society and economy. Therefore, it is necessary to evaluate the fire risk in the area in order to prevent and reduce the occurrence of fire accidents and provide references for fire departments and city construction and management departments in terms of disaster prevention and reduction measures.
Disclosure of Invention
In order to solve the vacancy and the deficiency of the regional fire risk assessment technology, the invention aims to provide a method for drawing a fire risk grade distribution map, which is used for providing data support and reference for disaster prevention and reduction by predicting the fire risk distribution situation and taking corresponding preventive measures in advance.
In order to realize the task, the invention adopts the following technical scheme:
a method for drawing a fire risk level distribution map comprises the following steps:
selecting a city for drawing a fire grade distribution diagram, and collecting fire influence factor data of historical fire occurrence places and fire condition data of the historical fire occurrence places and other areas in the city to establish a fire information distribution condition data set; the fire influencing factor data comprises environmental information, hazard source and meteorological information; the fire condition data of the historical fire point place comprises the size of the fire area, property loss and casualty conditions;
marking and extracting various landforms in the region in the city to obtain grid data, wherein the grid data is in a form that each longitude and latitude coordinate corresponds to one landform;
preprocessing a fire information distribution condition data set by using a data cleaning and feature extraction method;
the method comprises the steps that preprocessed fire information distribution condition data are concentrated, environmental information, hazard sources and meteorological information of a historical fire point place are used as independent variables, fire area size, property loss and casualty conditions are used as dependent variables, and a machine learning algorithm is adopted for training to obtain a fire area size prediction model, a property loss prediction model and a casualty condition prediction model; calculating the grid data by using the prediction models to obtain the fire area size, the property loss and the casualty condition prediction value corresponding to each grid;
determining the weight of each predicted value for the obtained predicted values of the fire area, the property loss and the casualty condition, calculating a final risk value, and performing normalization processing and grading;
and drawing a fire risk grade distribution diagram based on the grid data and the divided grades.
Further, the environment information specifically comprises a landform, population density and building density; the dangerous sources specifically comprise flammable and explosive articles, gas stations, chemical plants and electric facilities; the meteorological information specifically comprises temperature, humidity, wind speed and rainfall; to ensure accuracy, the grids should be as small as possible, so that each grid contains only one type of landform; all grid division and longitude and latitude coordinate conversion are processed according to the consistency principle.
Further, the marking and extracting various types of landforms in the region in the city to obtain grid data includes:
obtaining a tile map of a city, carrying out picture marking, and obtaining a training set; marking partial tile graphs by using a marking tool, wherein the marking types are residential areas, forests, rivers, oceans, farmlands, lakes and roads, each marked picture corresponds to one xml file, and the obtained xml files are analyzed in batches to obtain a training set required by a training model;
downloading an open-source neural network framework Darknet53, compiling the open-source neural network framework Darknet53, adding a training set and setting training weight after modifying a parameter file, and storing a trained model;
and marking all the tile graphs and extracting information by using the trained model.
Further, the preprocessing the fire information distribution data set by using the data cleaning and feature extraction method includes:
judging whether multiple collinearity exists between the input variables by utilizing multiple collinearity test; in order to eliminate multiple collinearity, a principal component analysis method is utilized, and on the premise that original information is kept as much as possible, data dimensionality is reduced, and the correlation degree and redundancy of data information are reduced; for the non-data volume, directly assigning values by adopting an expert experience method; and (4) carrying out normalization processing on the data by adopting a maximum normalization method and a minimum normalization method, so that the data range of each data is scaled to be between [0,1 ].
Further, the machine learning algorithm adopts a multiple linear regression algorithm to process data, and a model expression of the machine learning algorithm is as follows:
y＝β 0 +β 1 Z 1 +β 2 Z 3 +…+β k Z k +ε
in the formula: z is a linear or branched member k Is the kth principal component; beta is a 0 ，β 1 ，…，β k Is a regression coefficient; epsilon is a random error term; the least square method is adopted in the fitting mode of the multivariate linear regression model, so that the sum of squares of errors is minimum; the model evaluation index adopts the root mean square error.
Further, the determining the weight of each predicted value, calculating a final risk value, and performing normalization processing and grade division includes:
setting weights for three factors of the size of the fire area, the property loss and the casualty condition predicted value by utilizing an analytic hierarchy process, specifically establishing a comparison matrix, obtaining a characteristic value and a characteristic vector, judging whether the consistency test is passed or not by calculating a consistency index, a random consistency index and a consistency ratio, and if the consistency test is passed, determining the characteristic vector value as the weight corresponding to each factor;
and multiplying the three predicted values corresponding to each grid by the corresponding weight values to calculate a final risk value, and carrying out normalization processing on the data, wherein the numerical value is set to be low risk in the interval of [0, 0.3), the numerical value is set to be medium risk in the interval of [0.3, 0.5), the numerical value is set to be medium risk in the interval of [0.5, 0.75), and the numerical value is set to be high risk in the interval of [0.75,1 ].
Further, the drawing a fire risk level distribution map based on the grid data and the divided levels includes:
selecting and displaying an administrative range by utilizing attribute data according to boundary coordinates of a city, establishing a layer, drawing a boundary by utilizing an editing tool in ArcMap software by combining aerial images under the same coordinate system, and cutting; and importing the txt text containing all the data into ArcMap software, designating X fields and Y fields in the txt document as longitude and latitude fields, classifying all the data points according to the grade division and exporting the data points to the map layer to obtain a fire risk grade distribution map.
Compared with the prior art, the invention has the following technical characteristics:
1. according to the invention, relevant influence factors causing fire hazard are fully considered, the relation between fire hazard risk distribution and multi-source data is analyzed by combining the characteristic information of multi-source data such as different geographies, meteorology and regional landscape, and the drawn fire hazard risk grade distribution map is accurate and effective; the fire risk prediction is carried out based on the distribution diagram, effective decision suggestions can be provided for city construction management departments and fire departments, and the distribution diagram has important practical application value.
2. According to the method, a YOLOv3 automatic labeling algorithm, a mathematical statistics and a machine learning related algorithm are adopted, a multiple linear regression model is constructed through data processing, mining and model training, and a fire risk prediction model is obtained, so that the model is more accurately established.
Drawings
Fig. 1 is a schematic flow chart of a method for drawing a fire risk level distribution diagram according to the present invention.
Detailed Description
In order to clearly recognize the fire condition and the same standard, a fire risk grade distribution diagram is drawn, the fire risk distribution condition is known through the distribution diagram, the fire condition is predicted, and corresponding preventive measures are carried out, such as fire risk investigation and the like, in advance to prevent the fire from happening.
Referring to fig. 1, the method for drawing a fire risk level distribution map provided by the invention comprises the following steps:
step 1, constructing a fire information distribution condition data set
Selecting a city for drawing a fire grade distribution diagram, and collecting fire influence factor data of historical fire occurrence places and fire condition data of the historical fire occurrence places and other areas in the city to establish a fire information distribution condition data set; the fire impact factor data includes environmental information, hazard source, and weather information, wherein:
the environmental information specifically comprises landform, population density and building density; the dangerous sources specifically comprise flammable and explosive articles, gas stations, chemical plants and electric facilities; the meteorological information specifically comprises temperature, humidity, wind speed and rainfall; the fire condition data of the historical fire point occurrence place comprises the size of a fire area, property loss and casualty conditions.
The population density and building density data are obtained from related departments, and the data are spatially processed by utilizing spatial interpolation methods such as a surface interpolation method, a point interpolation method, a geographic statistics method and the like; the meteorological information is the average of meteorological elements in the last 15 days. Particularly, the fire influence factor data and the fire condition data are counted according to grids, in order to ensure accuracy, the grids are as small as possible, each grid only contains one type of landform as far as possible, and meteorological value errors are small. All the grid division and longitude and latitude coordinate conversion related by the invention are processed according to the consistency principle.
In this embodiment, a city is selected as a research object, and environmental information, hazard sources, and weather information of all fire occurrence points and all areas in the city in the last decade are collected. The data are acquired by means of fire department historical data, field research, weather monitoring station assistance and the like. And acquiring some uncovered areas by an interpolation method, and dividing the acquired data according to area grids.
Step 2, acquiring topographic and geomorphic data
At present, information extraction research aiming at regional landforms is less, so that the method utilizes a yoloV3 picture marking algorithm to mark and extract various landforms in the region to obtain grid data. Since the occurrence of fire is closely related to the landform, the information of the landform of the region needs to be obtained, YOLOv3 is a target monitoring network, and the automatic marking of the learned object is realized mainly by learning a large number of marked pictures. The method comprises the following specific steps:
and 2.1, obtaining a tile map. And (3) obtaining a tile map of the city by using water through map injection downloading software, wherein the picture size is 256 × 256.
And 2.2, carrying out picture marking by using labelimg software to obtain a training set. And (3) marking part of the tile map by using a marking tool labellimg, wherein the marking types are residential areas, forests, rivers, oceans, farmlands, lakes and roads, each marked picture corresponds to one xml file, and the obtained xml files are analyzed in batches to obtain a training set required by the training model.
And 2.3, deploying the model environment. Downloading an open-source neural network framework Darknet53 in a linux environment, compiling the open-source neural network framework Darknet53, adding a training set and setting training weight after modifying a parameter file, and storing a trained model; in this embodiment, the training weight is darknet53.Conv.74.
And 2.4, marking all the tile graphs and extracting information by using the trained model. Each tile map corresponds to one longitude and latitude and also correspondingly contains terrain type information, so that the generated grid data form is that each longitude and latitude coordinate corresponds to one terrain.
With reference to historical fire point data, the grid data is only missing the last three items. The data that needs to be collected are shown in table 1.
Table 1 example of data collection
Step 3, preprocessing the fire information distribution condition data set by using a data cleaning and feature extraction method
Step 3.1, data preprocessing
As the data collected in the fire information distribution condition data set contains various types, and for the non-data quantity, the assignment is directly carried out by adopting an expert experience method. Because the variable dimensions are not consistent and influence the research result, the data are normalized by adopting the maximum normalization method and the minimum normalization method respectively, so that the data range of each variable is scaled to [0,1], and the expression is as follows:
in the formula, X and Y are numerical values before and after normalization respectively; x max And X min The maximum and minimum values of the sample, respectively.
And 3.2, judging whether multiple collinearity exists among the data by utilizing multiple collinearity detection.
Multicollinearity means that the data in the model distort model estimation or are difficult to estimate accurately due to the existence of a high correlation relationship, and is evaluated by a variance expansion factor, wherein the formula is as follows:
in the formula:
is X j The regression coefficient R2 for all data indicates that no collinearity exists at all if VIF = 1; if 1 < VIF < 5, it means that multiple collinearity hardly exists; if 5 ≦ VIF < 10, multiple collinearity may be present; if VIF ≧ 10, it indicates that severe multicollinearity exists.
Step 3.3, feature extraction
In order to eliminate multiple collinearity, a principal component analysis method is utilized, and on the premise that original information is kept as far as possible, data dimensionality is reduced, the correlation degree and redundancy of data information are reduced, and modeling complexity is reduced.
Step 4, building a fire risk prediction model
The method comprises the steps of centralizing preprocessed fire information distribution condition data, using fire influence factor data of historical fire site places, namely environment information, hazard sources and meteorological information as independent variables, using fire condition data, namely fire area size, property loss and casualty condition as dependent variables, and adopting prediction algorithms such as multivariate linear regression, gray linear regression, support vector regression, decision trees, bayes and the like in the field of machine learning to train so as to obtain a fire area size prediction model, a property loss prediction model and a casualty condition prediction model.
Because the multiple linear regression model is mainly used for judging the relationship between a plurality of predictive variables and the predictive index, the multiple linear regression algorithm is adopted to process data in the embodiment, and the model expression is as follows:
y＝β 0 +β 1 Z 1 +β 2 Z 3 +…+β + Z k +ε
in the formula: z k Is the kth principal component; beta is a beta 0 ，β 1 ，…，β k Is a regression coefficient; ε is a random error term. The fitting mode of the multiple linear regression model adopts a common least square method, namely suitable beta is found 0 ，β 1 ，…，β k So that the sum of squares of errors
A minimum is reached where n is the sample size. The model evaluation index adopts Root Mean Square Error (RMSE), the RMSE is used for measuring the deviation between a predicted value and a true value, the smaller the value is, the higher the prediction precision is, and the expression is as follows:
in the formula:
is the predicted value of the model at the moment t; y (t) is the real value of the model at the moment t; n is the sample size.
And (3) after a fire area size prediction model, a property loss prediction model and a casualty condition prediction model are established by utilizing a multiple linear regression algorithm, calculating all grid data obtained in the step (2) by utilizing each prediction model to obtain a fire area size, property loss and a casualty condition prediction value corresponding to each grid.
Step 5, risk grade division is carried out based on the predicted value
And (5) determining the weight of each predicted value by using methods such as a manual judgment method, a sequencing method, a ratio method, a pair comparison method and the like for the predicted values of the fire area size, the property loss and the casualty condition obtained in the step (4), calculating a final risk value, and performing normalization processing and grade division.
In this embodiment, the method of analytic hierarchy process is used to set weights for the three factors of the area of fire, the property loss and the casualty, specifically including establishing a comparison matrix, obtaining a characteristic value and a characteristic vector, and calculating a consistency index
Random consistency index RI and consistency ratio
And judging whether the consistency test is passed, if the consistency test is passed, the characteristic vector value is the weight corresponding to each factor.
And multiplying the three predicted values corresponding to each grid by the corresponding weight values to calculate a final risk value, and carrying out normalization processing on the data, wherein the numerical value is set to be low risk in the interval of [0, 0.3), the numerical value is set to be medium risk in the interval of [0.3, 0.5), the numerical value is set to be medium risk in the interval of [0.5, 0.75), and the numerical value is set to be high risk in the interval of [0.75,1 ].
And 6, drawing a fire risk grade distribution map.
And drawing a fire risk grade distribution map by using ArcMap software. Selecting and displaying an administrative range by utilizing attribute data according to boundary coordinates of the city, establishing a layer, drawing a boundary by utilizing an editing tool in ArcMap by combining aerial images under the same coordinate system, and cutting; and importing the txt text containing all the data into an ArcMap, designating X fields and Y fields in the txt document as longitude and latitude fields by using a Display XY data \8230menu, classifying all the data points according to the grade division in the step 5, and exporting the data points to a map layer to obtain a fire risk grade distribution map.
The above embodiments are only used to illustrate the technical solutions of the present application, and not to limit the same; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; such modifications and substitutions do not depart from the spirit and scope of the embodiments of the present application, and they should be construed as being included in the present application.
Claims (1)
1. A method for drawing a fire risk level distribution map is characterized by comprising the following steps:
selecting a city for drawing a fire grade distribution map, and collecting fire influence factor data of historical fire point places and fire condition data of the historical fire point places in the city to establish a fire information distribution condition data set; the fire influencing factor data comprises environmental information, hazard sources and meteorological information; the fire condition data of the historical fire point place comprises the size of the fire area, property loss and casualty conditions;
marking and extracting various landforms in the region in the city to obtain grid data, wherein the grid data is in a form that each longitude and latitude coordinate corresponds to one landform;
preprocessing a fire information distribution condition data set by using a data cleaning and feature extraction method;
the method comprises the steps that preprocessed fire information distribution condition data are concentrated, environmental information, danger sources and meteorological information of a historical fire point place are used as independent variables, fire area size, property loss and casualty conditions are used as dependent variables, and a machine learning algorithm is adopted for training to obtain a fire area size prediction model, a property loss prediction model and a casualty condition prediction model; calculating the grid data by using the prediction models to obtain the fire area size, the property loss and the casualty condition prediction value corresponding to each grid;
determining the weight of each predicted value according to the obtained predicted values of the fire area size, the property loss and the casualty condition, calculating a final risk value, and performing normalization processing and grading;
drawing a fire risk grade distribution diagram based on the grid data and the divided grades;
the environment information specifically comprises landform, population density and building density; the dangerous sources specifically comprise flammable and explosive articles, gas stations, chemical plants and electric facilities; the meteorological information specifically comprises temperature, humidity, wind speed and rainfall; in order to ensure the accuracy, each grid only contains one type of landform; all grid division and longitude and latitude coordinate conversion are processed according to a consistency principle;
the marking and extracting of various types of landforms in the region in the city to obtain grid data comprises the following steps:
obtaining a tile map of a city, carrying out picture marking, and obtaining a training set; marking a part of tile graphs by using a marking tool, wherein the marking types are residential areas, forests, rivers, oceans, farmlands, lakes and roads, each marked picture corresponds to an xml file, and the obtained xml files are analyzed in batches to further obtain a training set required by a training model;
downloading an open-source neural network framework Darknet53, compiling the open-source neural network framework Darknet53, adding a training set and setting training weights after modifying a parameter file, and storing a trained model;
marking all the tile graphs and extracting information by using the trained model;
the method for preprocessing the fire information distribution condition data set by utilizing the data cleaning and feature extraction method comprises the following steps:
judging whether multiple collinearity exists between the input variables by utilizing multiple collinearity test; in order to eliminate multiple collinearity, a principal component analysis method is utilized, the data dimensionality is reduced, and the correlation degree and the redundancy of data information are reduced; for the non-data volume, directly assigning values by adopting an expert experience method; carrying out normalization processing on the data by adopting a maximum normalization method and a minimum normalization method, so that the data range of each data is scaled to be between [0,1 ];
the machine learning algorithm adopts a multiple linear regression algorithm to process data, and the model expression is as follows:
y＝β 0 +β 1 Z 1 +β 2 Z 3 +…+β k Z k +ε
in the formula: z k Is the kth principal component; beta is a beta 0 ,β 1 ,…,β k Is a regression coefficient; epsilon is a random error term; the least square method is adopted in a fitting mode of the multivariate linear regression model, so that the sum of squares of errors is minimum; the model evaluation index adopts a root-mean-square error;
determining the weight of each predicted value, calculating a final risk value, and performing normalization processing and grade division, wherein the method comprises the following steps:
setting weights for three factors of the size of the area of the fire, the property loss and the casualty condition predicted value by using an analytic hierarchy process, specifically comprising establishing a comparison matrix, obtaining a characteristic value and a characteristic vector, judging whether consistency inspection is passed or not by calculating a consistency index, a random consistency index and a consistency ratio, and if the consistency inspection is passed, determining the characteristic vector value as the weight corresponding to each factor;
multiplying the three predicted values corresponding to each grid by corresponding weight values to calculate final risk values, and carrying out normalization processing on data, wherein the numerical values are set to be low risk in a range of [0, 0.3), the numerical values are set to be medium risk in a range of [0.3, 0.5), the numerical values are set to be medium risk in a range of [0.5, 0.75), and the numerical values are set to be high risk in a range of [0.75,1 ];
the drawing of the fire risk level distribution map based on the grid data and the divided levels includes:
selecting and displaying an administrative range by utilizing attribute data according to boundary coordinates of a city, establishing a layer by combining aerial images under the same coordinate system, drawing a boundary by utilizing an editing tool in ArcMap software, and cutting; and importing the txt text containing all the data into ArcMap software, designating X fields and Y fields in the txt document as longitude and latitude fields, classifying all the data points according to the grade division and exporting the data points to the map layer to obtain a fire risk grade distribution map.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011077194.7A CN112308292B (en) | 2020-10-10 | 2020-10-10 | Method for drawing fire risk grade distribution map |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011077194.7A CN112308292B (en) | 2020-10-10 | 2020-10-10 | Method for drawing fire risk grade distribution map |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112308292A CN112308292A (en) | 2021-02-02 |
| CN112308292B true CN112308292B (en) | 2023-01-20 |
Family
ID=74489530
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011077194.7A Active CN112308292B (en) | 2020-10-10 | 2020-10-10 | Method for drawing fire risk grade distribution map |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112308292B (en) |
Families Citing this family (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112991670A (en) * | 2021-02-04 | 2021-06-18 | 西安美格智联软件科技有限公司 | Fire-fighting dangerous area classification management and control method and system, storage medium and processing terminal |
| CN112884324A (en) * | 2021-02-21 | 2021-06-01 | 深圳数研锦瀚智慧科技有限公司 | Forest fire danger monitoring method and device and storage medium |
| CN113033391B (en) * | 2021-03-24 | 2022-03-08 | 浙江中辰城市应急服务管理有限公司 | Fire risk early warning research and judgment method and system |
| CN113554281A (en) * | 2021-07-02 | 2021-10-26 | 北京淇瑀信息科技有限公司 | Grid-based user business risk analysis method and device and electronic equipment |
| CN113793066B (en) * | 2021-09-30 | 2022-04-01 | 成都安讯智服科技有限公司 | Item position aggregation method, system, terminal and medium based on risk analysis |
| CN113902963B (en) * | 2021-12-10 | 2022-06-17 | 交通运输部公路科学研究所 | Method and device for evaluating fire detection capability of tunnel |
| CN114298392B (en) * | 2021-12-23 | 2023-05-12 | 电子科技大学 | Fire wire longitude and latitude prediction method for forest fire |
| CN115273438B (en) * | 2022-07-05 | 2023-05-02 | 广东远景信息科技有限公司 | Forest intelligent fireproof method, device, system, equipment and medium based on 5G |
| CN115147024B (en) * | 2022-09-05 | 2022-12-13 | 杭州元声象素科技有限公司 | Gridding dangerous case processing method and system of geographic weighted regression |
| CN117592789B (en) * | 2024-01-18 | 2024-04-16 | 山东金桥保安器材有限公司 | Power grid environment fire risk assessment method and equipment based on time sequence analysis |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108009351A (en) * | 2017-11-30 | 2018-05-08 | 国网湖南省电力有限公司 | The distribution drawing drawing method of lightning stroke trip risk |
| CN109064050A (en) * | 2018-08-17 | 2018-12-21 | 公安部沈阳消防研究所 | Multiple linear regression Fire risk assessment method based on big data |
| CN111611524A (en) * | 2020-04-17 | 2020-09-01 | 北京市燃气集团有限责任公司 | Gas risk assessment and safety supervision resource matching method and device |
| CN111737651A (en) * | 2020-06-22 | 2020-10-02 | 黄河勘测规划设计研究院有限公司 | Spatial gridding drought disaster risk assessment method and system based on multi-source data |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108280553B (en) * | 2018-02-24 | 2020-10-02 | 中山大学 | Mountain torrent disaster risk zoning and prediction method based on GIS-neural network integration |
| CN109858647B (en) * | 2018-12-21 | 2021-07-27 | 河海大学 | Regional flood disaster risk evaluation and estimation method coupled with GIS and GBDT algorithm |
-
2020
- 2020-10-10 CN CN202011077194.7A patent/CN112308292B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108009351A (en) * | 2017-11-30 | 2018-05-08 | 国网湖南省电力有限公司 | The distribution drawing drawing method of lightning stroke trip risk |
| CN109064050A (en) * | 2018-08-17 | 2018-12-21 | 公安部沈阳消防研究所 | Multiple linear regression Fire risk assessment method based on big data |
| CN111611524A (en) * | 2020-04-17 | 2020-09-01 | 北京市燃气集团有限责任公司 | Gas risk assessment and safety supervision resource matching method and device |
| CN111737651A (en) * | 2020-06-22 | 2020-10-02 | 黄河勘测规划设计研究院有限公司 | Spatial gridding drought disaster risk assessment method and system based on multi-source data |
Non-Patent Citations (5)
| Title |
|---|
| 住宅建筑火灾财产损失影响因素及其实证研究;屈丽娟;《火灾科学》;20070225(第01期);全文 * |
| 回归分析中多重共线性的诊断与处理;魏红燕;《周口师范学院学报》;20190315(第02期);第11-15页 * |
| 基于ArcGIS矿产资源规划数据库建设方法及问题探讨;魏泽权;《贵州地质》;20130615(第02期);全文 * |
| 基于主动深度学习的高光谱影像分类;程圆娥;《计算机工程与应用》;20170922;第192-196页 * |
| 森林火灾面积预测在保护生态环境中的应用;刘旭菲;《资源节约与环保》;20180125(第01期);第99-100页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112308292A (en) | 2021-02-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112308292B (en) | Method for drawing fire risk grade distribution map | |
| Klouček et al. | How does data accuracy influence the reliability of digital viewshed models? A case study with wind turbines | |
| CN112070286A (en) | Rainfall forecast early warning system for complex terrain watershed | |
| CN110646867A (en) | Urban drainage monitoring and early warning method and system | |
| Cheng et al. | Statistical downscaling of hourly and daily climate scenarios for various meteorological variables in south-central Canada | |
| CN116205541B (en) | Method and device for evaluating influence of local pollution source on environmental air quality | |
| CN110738354A (en) | Method and device for predicting particulate matter concentration, storage medium and electronic equipment | |
| CN108764527B (en) | Screening method for soil organic carbon library time-space dynamic prediction optimal environment variables | |
| Samadi et al. | Comparison of general circulation models: methodology for selecting the best GCM in Kermanshah Synoptic Station, Iran | |
| CN115994685A (en) | Method for evaluating current situation of homeland space planning | |
| CN114048944A (en) | Estimation method for people to be evacuated and houses to be damaged under rainstorm induced geological disaster | |
| EP4287214A1 (en) | Information processing device, water resource managing method, information processing method, and recording medium | |
| CN116822185A (en) | Daily precipitation data space simulation method and system based on HASM | |
| Taubenböck et al. | Remote sensing—An effective data source for urban monitoring | |
| CN113506371B (en) | Street scale climatic diagram drawing method and device, electronic equipment and storage medium | |
| JP7224133B2 (en) | Infrastructure equipment inspection support system, infrastructure equipment inspection support method and program | |
| Pappa et al. | Analysis of the ZR relationship using X-Band weather radar measurements in the area of Athens | |
| Schlager et al. | Generation of high-resolution wind fields from the WegenerNet dense meteorological station network in southeastern Austria | |
| CN110019167A (en) | Long-term new forms of energy resource data base construction method and system in one kind | |
| CN114818857A (en) | Deep snow fusion method | |
| CN117708551B (en) | Flood disaster influence assessment method and system based on double-precision GDP data distribution | |
| CN117610434B (en) | Artificial intelligence fused drought index reconstruction method and computer readable medium | |
| CN113807724B (en) | Site selection method for slag disposal site based on comprehensive risk evaluation | |
| Zappa et al. | The Changjiang flood forecasting assistance project | |
| Zahran et al. | Assessing of Domain Change Sensitivity for Regional Climate Model simulation (Reg-Cm4. 3) at Blue Nile Basin |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |