CN115330254A - Mountain torrent comprehensive risk early warning method, system and storage medium - Google Patents

Abstract

The application relates to a method, a system and a storage medium for mountain torrent comprehensive risk early warning, which belong to the field of disaster monitoring, wherein the method comprises the steps of acquiring attribute information of a target small watershed based on a preset mountain torrent risk basic attribute data set; classifying the attribute information to obtain attribute factors; the attribute factors comprise rainfall factors, terrain factors and social factors; constructing a correlation coefficient matrix for the attribute factors; performing principal component analysis on the attribute factors based on the correlation coefficient matrix to obtain a plurality of key factors; normalizing the plurality of key factors to obtain the weight of the risk factor; calculating a risk index based on the risk factor weight; and determining the early warning grade corresponding to the risk index based on a preset risk index range. The application has the effect of effectively improving the accuracy of the early warning of the torrential rain and torrential flood disasters in the small watershed.

Description

A method, system and storage medium for comprehensive risk warning of mountain torrents

technical field

The present application relates to the field of disaster monitoring, in particular to a method, system and storage medium for comprehensive risk early warning of mountain torrents.

Background technique

Early warning of rainstorm and mountain torrent disasters in small watersheds is one of the most effective means to prevent mountain torrent disasters. If early warning information is sent out in time when or before a disaster occurs, casualties and property losses can be effectively reduced. However, the formation process of torrential rain and mountain torrents in small watersheds is complicated. The method of mountain torrent forecasting and early warning in the prior art is mainly based on the analysis of 24-hour rainfall forecast data. Determining the areas where mountain torrent disasters may occur and the early warning level effectively extended the early warning period.

In view of the prior art mentioned above, the applicant believes that because the current forecast of local short-duration rainfall is often inaccurate, that is, the forecast accuracy is low, and under the condition of inaccurate forecast accuracy, it will cause the early warning period in the prior art. Accuracy is greatly reduced.

Contents of the invention

In order to effectively improve the accuracy of early warning of torrential rain and mountain torrent disasters in small watersheds, the application provides a method, system and storage medium for comprehensive risk early warning of torrents.

In the first aspect, a kind of mountain torrent comprehensive risk early warning method provided by the application adopts the following technical scheme:

A comprehensive risk early warning method for mountain torrents, comprising:

Obtain the attribute information of the target small watershed based on the preset mountain torrent risk basic attribute data set;

Classifying the attribute information to obtain attribute factors; the attribute factors include rainfall factors, terrain factors and social factors;

Constructing a correlation coefficient matrix for the attribute factors;

Based on the correlation coefficient matrix, performing principal component analysis on the attribute factors to obtain several key factors;

Normalize several key factors to obtain risk factor weights;

calculating a risk index based on the risk factor weights;

Based on the preset risk index range, an early warning level corresponding to the risk index is determined.

By adopting the above-mentioned technical scheme, firstly, principal component analysis is performed on the attribute factors to obtain key factors, and then the risk factor indicators are obtained based on the key factors, and the risk index is calculated based on the risk factor indicators. Effectively improve the accuracy of early warning and small watershed rainstorm mountain torrent disaster early warning accuracy.

Optionally, the torrent risk basic attribute data set includes vector data of a rainstorm atlas, small watershed attribute data and forecast rainfall data;

Before classifying the attribute information, it includes:

Combining the vector data of the rainstorm atlas with the attribute data of the small watershed;

Based on the inverse distance weighting method, the forecast rainfall data is interpolated into grid data with a preset resolution.

By adopting the above technical scheme, the vector data of the rainstorm atlas is historical statistical data, and the vector data of the rainstorm atlas is combined with the attribute data of small watersheds to judge the risk ability of small watersheds to resist mountain torrents. The anti-risk ability is strong, and the small watershed that does not often produce heavy rain is judged to have a weak anti-risk ability, that is, the factor of anti-risk ability is added to the attribute data of the small watershed, which is conducive to more accurate judgments on subsequent disaster warnings.

Optionally, based on the correlation coefficient matrix, principal component analysis is performed on the attribute factors to obtain several key factors, including:

Based on the preset principal component analysis model and the preset certainty coefficient model, the factor sensitivity of the attribute factor is calculated; according to the factor sensitivity, the key factor is determined.

By adopting the above technical scheme, the principal component analysis model is used to convert multiple indicators into a few comprehensive indicators, and the certainty coefficient model is used to realize the quantification of complex multi-factor data. The determination of key factors is conducive to the analysis of the impact of mountain torrent disasters factor, and pave the way for the subsequent quantification of mountain torrent disasters.

Optionally, the calculation of the factor sensitivity of the attribute factors based on the preset principal component analysis model and the preset certainty coefficient model includes:

Calculating the first ratio of the number of disasters with mountain torrent disasters and the area of the preset target area under each attribute factor;

Calculating the second ratio of the number of disasters under each attribute factor to the preset hill area;

Calculate the certainty coefficients of the rainfall factor, the terrain factor and the social factor respectively based on the first ratio, the second ratio and the preset certainty coefficient model formula;

Calculate the factor sensitivity of the attribute factor based on the preset principal component analysis model and the certainty coefficient.

By adopting the above technical scheme, the certainty coefficient model is a probability function, which is used to analyze the sensitivity of various attribute factors affecting the occurrence of mountain torrent disasters. Interval quantification.

Optionally, the certainty coefficient model formula is:

Wherein, the PPa is the first ratio; PPs is the second ratio; CF is the coefficient of certainty.

By adopting the above technical scheme, the certainty coefficient model is a probability function, which is used to analyze the sensitivity of various attribute factors affecting the occurrence of mountain torrent disasters, which is conducive to paving the way for the subsequent determination of key factors.

Optionally, the calculating the risk index based on the risk factor weights includes:

Calculating the risk factor weights based on a combination of subjective and objective methods;

The risk index is obtained by adding the weights of several risk factors.

By adopting the above technical scheme, the risk index is the product of the factor weight and the risk factor index corresponding to the weight, so the impact on the mountain torrent disaster is quantified, and the accuracy of the small watershed rainstorm torrent disaster warning is effectively improved.

Optionally, after determining the warning level corresponding to the risk index within the preset warning level range, it includes:

Mark the target area corresponding to the risk warning in the preset area map.

By adopting the above-mentioned technical solution, marking the risk early warning of the target area is beneficial to visually see the risk of mountain torrent disasters in different positions in the area and play an early warning role.

In the second aspect, a kind of mountain torrent comprehensive risk early warning system provided by the application adopts the following technical scheme:

A comprehensive risk early warning system for mountain torrents, comprising:

An acquisition module, the acquisition module is used to acquire the attribute information of the target small watershed based on the preset flash flood risk basic attribute data set;

A classification module, the classification module is used to classify the attribute information to obtain attribute factors;

A building block, the building block is used to build a correlation coefficient matrix for the attribute factors;

An analysis module, the analysis module is used to perform principal component analysis on the attribute factors based on the correlation coefficient matrix to obtain several key factors;

A processing module, the processing module is used to normalize several key factors to obtain risk factor indicators;

A calculation module, the calculation module is used to calculate the risk index based on the risk factor index;

A determination module, configured to determine the warning level corresponding to the risk index based on the preset risk index range.

By adopting the above technical scheme, firstly, the principal component analysis is performed on the attribute factors through the analysis module to obtain the key factors, and then the risk factor indicators are obtained based on the key factors through the processing module, and then the risk index is calculated based on the risk factor indicators through the calculation module, that is, realized The quantification of early warning of rainstorm and mountain torrent disasters in small watersheds has effectively improved the accuracy of early warning and the accuracy of early warning of rainstorm and torrent disasters in small watersheds.

In the third aspect, a storage medium provided by the present application adopts the following technical solution:

A computer-readable storage medium, wherein a computer program is stored in the computer-readable storage medium, and when the computer program is loaded and executed by a processor, the above-mentioned integrated mountain torrent risk early warning method is adopted.

By adopting the above-mentioned technical solution, by generating a computer program from the above-mentioned comprehensive risk warning method for mountain torrents, and storing it in a computer-readable storage medium, so as to be loaded and executed by a processor, the computer-readable storage medium facilitates the readability of the computer program and storage.

In summary, the present application has at least one of the following beneficial technical effects:

1. First, conduct principal component analysis on attribute factors to obtain key factors, then obtain risk factor indicators based on key factors, and calculate risk index based on risk factor indicators, that is, realize the quantification of early warning of rainstorm and mountain torrent disasters in small watersheds, and effectively improve early warning Accuracy and accuracy of small watershed rainstorm and mountain torrent disaster warning.

2. The vector data of the rainstorm atlas is historical statistical data. Combining the vector data of the rainstorm atlas with the attribute data of small watersheds is used to judge the risk ability of small watersheds against mountain torrents, that is, adding the anti-risk ability to the attribute data of small watersheds Factors are conducive to more accurate judgments on subsequent disaster warnings.

3. The risk index is the product of the factor weight and the risk factor index corresponding to the weight, so the impact on mountain torrent disasters has been quantified, effectively improving the accuracy of early warning of torrential rain and mountain torrent disasters in small watersheds.

Description of drawings

FIG. 1 is an overall flowchart of a method for comprehensive risk early warning of mountain torrents according to an embodiment of the present application.

FIG. 2 is a flow chart of calculating the factor sensitivity of attribute factors based on a preset principal component analysis model and a preset certainty coefficient model in a method for comprehensive risk warning of torrents according to an embodiment of the present application.

Detailed ways

The embodiment of the present application discloses a comprehensive risk early warning method for mountain torrents.

Referring to Figure 1, a comprehensive risk early warning method for mountain torrents includes:

S100. Obtain attribute information of the target small watershed based on the preset torrent risk basic attribute data set.

In this embodiment, the mountain torrent risk basic attribute data set includes mountain torrent disaster risk area geographic data, hydrological data and historical mountain torrent disaster data. That is, the geographical data, hydrological data and historical torrent disaster data of the mountain torrent disaster risk area are collected first, and then the geographic data, hydrological data and historical torrent disaster data are organized into a small watershed torrent risk basic attribute data set. In this embodiment, the small watershed torrent risk The basic attribute data set takes a small watershed less than 200 square kilometers as a unit. The attribute information of the target small watershed includes several attribute information fields, and in the actual implementation, there are 77 attribute information fields.

S200. Classify the attribute information to obtain attribute factors; the attribute factors include rainfall factors, terrain factors and social factors.

Attribute factors include rainfall factors, terrain factors and social factors. In this embodiment, the 77 attribute information fields of each small watershed are divided into 3 categories, namely, rainfall factors, terrain factors and social factors, wherein rainfall factors account for 59 There are 15 attribute information fields for terrain factors, and 3 attribute information fields for social factors. It should be noted that, compared with the existing classification of attribute information, social factors are introduced in this embodiment, and comprehensive early warning and quantification of flash flood risk are carried out based on rainfall factors, terrain factors and social factors.

Specifically, the basic attribute data set of flash flood risk includes vector data of rainstorm atlas, attribute data of small watersheds and forecast rainfall data;

Before classifying attribute information, including:

S1. Combining the vector data of the rainstorm atlas with the attribute data of small watersheds.

The vector data of the rainstorm atlas are used to express the characteristics and laws of the temporal and spatial distribution of the statistical characteristics of the rainstorm. The vector data of the rainstorm atlas are historical statistical data. The vector data of the rainstorm atlas are combined with the attribute data of small watersheds to judge the ability of small watersheds to resist mountain torrents. Risk capacity, for example, if the frequency of rainstorm in a small watershed is higher than that of other small watersheds within the preset time range, the residents in this small watershed have formed habits and have taken measures to prevent heavy rains, so it is judged that the small watershed’s resistance The risk ability is stronger than other small watersheds, that is, the small watersheds that often produce heavy rains are judged to have strong anti-risk capabilities, and the small watersheds that do not often produce heavy rains are judged to be weak in anti-risk capabilities.

Adding the factor of anti-risk ability to the attribute data of small watersheds will help to make more accurate judgments on subsequent disaster early warnings.

S2. Based on the inverse distance weighting method, the forecast rainfall data is interpolated into grid data with a preset resolution.

The main idea of the inverse distance weighting method is the explicit assumption of inverse distance weight interpolation. The explicit assumption of inverse distance weight interpolation means that things that are closer to each other are more similar than things that are farther away from each other.

When the predicted value of an unmeasured predicted position is acquired, the measured value closest to the predicted position has a greater influence on the predicted value than the measured value farther from the predicted position. The inverse distance weighting method assumes that the measurement point corresponding to each measurement value has a local influence, and the local influence will decrease as the distance increases. The inverse distance weighting method assigns larger weights to points closer to the predicted position and less weight to points farther from the predicted position.

Inverse distance weighting method is a kind of spatial interpolation method. In addition, spatial interpolation methods also include kriging interpolation method, natural neighbor interpolation method, spline function interpolation method and radial basis function, etc.

The spatial resolution of the grid data in this embodiment is 5 kilometers by 5 kilometers. Spatial resolution is preset. The forecasted rainfall data is the forecasted rainfall data of the station in the next 24 hours.

Referring to FIG. 1 , S300 , construct a correlation coefficient matrix for attribute factors.

In this embodiment, the Pearson correlation coefficient method is used to construct the correlation coefficient matrix. The correlation coefficient matrix consists of the correlation coefficients subtracted from the columns of the matrix.

The degree of correlation between variables can be determined according to the value of the correlation coefficient.

Specifically, the value range and correlation degree of the correlation coefficient are shown in the following table: Ranges Relevance 0.00-0.19 very low correlation 0.20-0.39 low correlation 0.40-0.69 Moderately relevant 0.70-0.89 Highly correlated 0.90-1.00 extremely high correlation

Referring to FIG. 1 , S400 , based on the correlation coefficient matrix, perform principal component analysis on attribute factors to obtain several key factors.

Principal component analysis is a statistical method. Through orthogonal transformation, a group of variables that may be correlated is converted into a group of linearly uncorrelated variables, and the converted group of variables is called the principal component. Through principal component analysis of attribute factors, several key factors are obtained.

In this embodiment, after using the Pearson correlation coefficient method to construct the correlation coefficient matrix, the factors corresponding to the correlation coefficients with low correlation and extremely low correlation are eliminated, and then principal component analysis is performed on the remaining attribute factors to obtain several key factor.

Specifically, based on the correlation coefficient matrix, principal component analysis is performed on the attribute factors to obtain several key factors, including:

S410. Calculate the factor sensitivity of the attribute factor based on the preset principal component analysis model and the preset certainty coefficient model.

The certainty coefficient model is a probability function, which is used to analyze the sensitivity of each attribute factor that affects the occurrence of mountain torrent disasters. Using the certainty coefficient model to determine the sensitivity of attribute factors realizes the quantification of complex multi-attribute factor data in the same interval. The interval range of the same interval is [1, -1].

Specifically, referring to Figure 2, based on the preset principal component analysis model and the preset certainty coefficient model, the factor sensitivity of attribute factors is calculated, including:

S411. Calculate the first ratio of the number of disasters with flash flood disasters and the area of the preset target area under each attribute factor.

The number of disasters with flash flood disasters can be obtained through the database or through human input.

The target area area is the area area of the non-hill area. In the specific implementation, it is to calculate the first ratio of the number of disasters of mountain torrent disasters and the area of the preset target area under each attribute factor within the preset time range.

S412. Calculate a second ratio between the number of disasters under each attribute factor and the area of the preset hill area.

In a specific implementation, the second ratio between the number of disasters and the area of the preset hill area under each attribute factor within the preset time range is calculated.

S413. Based on the first ratio, the second ratio and the preset certainty coefficient model formula, respectively calculate the certainty coefficients of the rainfall factor, the terrain factor and the social factor.

S414. Calculate the factor sensitivity of the attribute factor based on the preset principal component analysis model and certainty coefficient.

Specifically, the formula of the certainty coefficient model is:

Among them, PPa is the first ratio; PPs is the second ratio; CF is the certainty coefficient.

Based on steps S411 to S413, an example is given. Taking the terrain factor in the attribute factor as an example, if in the terrain factor, within the time frame of 5 years, the number of mountain torrent disasters is 5, and the area of the target area is 20 square kilometers. If the area of the mound is 50 square kilometers, the first ratio PPa is 0.25, and the second ratio PPs is 0.1. At this time, PPa is greater than PPs, and the coefficient of certainty CF is (0.25-0.1)/0.25(1-0.1)=0.67.

S420. Determine key factors according to factor sensitivities.

According to the certainty coefficient model formula, the certainty coefficient calculation result is obtained, and the principal component analysis can be carried out according to the certainty coefficient calculation result and the correlation coefficient matrix.

In the specific implementation, the certainty coefficient obtained through the certainty coefficient model is the weight of the attribute factor. If the certainty coefficient of a certain attribute factor is greater than the preset coefficient threshold, the attribute factor is judged to be sensitive to mountain torrent disasters. factor, the key factor. For example, if the coefficient threshold is set to 0.06, and the certainty coefficient of the target factor is 0.08, since it is greater than the coefficient threshold, it is determined that the target factor is a sensitive factor.

The sensitivity of the factor is illustrated based on the rainfall factor. The rainfall factor includes the 6-hour average rainfall. If the 6-hour average rainfall is less than or equal to 25mm, the calculation result of the certainty coefficient is -0.83; between 25mm and 50mm , the calculation result is -0.59; between 50mm and 100mm, the calculation result is -0.2; between 100mm and 150mm, the calculation result is -0.04; when it is greater than 150mm, the calculation result is 0.24. From this, it can be concluded that the greater the 6-hour average rainfall, the higher the risk of flash flood disasters, and the 6-hour average rainfall is the key factor.

In the present embodiment, key factors include 6-hour average rainfall, 6-hour rainfall coefficient of variation, future rainfall, flood peak modulus, river length, river course gradient, family property, house and population, wherein family property, house and population are Social factors, 6-hour average rainfall, 6-hour rainfall variation coefficient, future rainfall and flood peak modulus are rainfall factors, and river length and river channel gradient are topographic factors.

Referring to FIG. 1 , S500 , perform normalization processing on several key factors to obtain risk factor weights.

Normalization processing is to uniformly map the data of key factors to the range of 0 to 1. Pave the way for the subsequent calculation of the risk index.

S600. Calculate the risk index based on the risk factor weights.

The risk index is the quantification of the early warning of mountain torrent disasters.

Specifically, based on the risk factor weights, the risk index is calculated, including:

S610. Calculate risk factor weights based on a combination of subjectivity and objectiveness.

Subjective and objective combination method Combination weighting method of subjective and objective combination, subjective and objective combination method often adopts multiplication or linear synthesis method, the current methods for determining the weight of index attributes can be divided into: subjective weighting method, objective weighting method and combination weighting 3 categories of law. Among them, the results of the subjective weighting method are relatively subjective. Although the objective weighting method has a mathematical theoretical basis, it does not consider the intention of the decision maker. The combination of subjective and objective methods is used to make up for the lack of single weighting. In this embodiment, the subjective-objective combination method is a method that combines the subjective weighting method and the entropy weighting method.

S620. Add up the weights of several risk factors to obtain a risk index.

Based on the number of certainty coefficients of the correlation coefficient matrix, there are several risk factor weights, and the risk index can be obtained by adding the weights of several risk factors of each risk factor.

Referring to FIG. 1 , S700, based on the preset risk index range, determine the warning level corresponding to the risk index.

In specific implementation, the warning levels and risk index ranges are shown in the table below: alert level Graphic color representation Risk Index Range may happen blue 0.45-0.6 more likely yellow 0.6-0.75 high possibility orange 0.75-0.9 very likely red 0.9-1

For example, if the risk index is 0.7, it is determined that the early warning level corresponding to the risk index is more likely.

Specifically, in the preset warning level range, after determining the warning level corresponding to the risk index, it includes:

S710. Mark the target area corresponding to the risk warning in the preset area map.

The regional map is preset, and each warning level corresponds to a color, and different warning levels have different colors.

The implementation principle of a comprehensive mountain torrent risk early warning method in the embodiment of the present application is as follows: first, principal component analysis is performed on the attribute factors to obtain key factors, and then the risk factor indicators are obtained based on the key factors, and the risk index is calculated based on the risk factor indicators. The quantification of early warning of rainstorm and mountain torrent disasters in small watersheds has effectively improved the accuracy of early warning and the accuracy of early warning of rainstorm and torrent disasters in small watersheds.

The embodiment of the present application also discloses a comprehensive risk warning system for mountain torrents.

A comprehensive risk early warning system for mountain torrents includes:

The acquisition module is used to obtain the attribute information of the target small watershed based on the preset flash flood risk basic attribute data set; the classification module is used to classify the attribute information to obtain attribute factors;

A building block, which is used to construct a correlation coefficient matrix for attribute factors;

Analysis module, the analysis module is used to conduct principal component analysis on attribute factors based on the correlation coefficient matrix to obtain several key factors;

A processing module, the processing module is used to normalize several key factors to obtain risk factor indicators;

Calculation module, the calculation module is used to calculate the risk index based on the risk factor index;

The determination module is configured to determine the warning level corresponding to the risk index based on the preset risk index range.

The implementation principle of a comprehensive mountain torrent risk early warning system in the embodiment of the present application is as follows: first, the principal component analysis is performed on the attribute factors through the analysis module to obtain the key factors, and then the risk factor indicators are obtained based on the key factors through the processing module, and then through the calculation module and based on The calculation of the risk index by the risk factor index realizes the quantification of the early warning of rainstorm and mountain torrent disasters in small watersheds, and effectively improves the accuracy of early warning and the accuracy of early warning of rainstorm and torrent disasters in small watersheds.

The embodiment of the present application also discloses a storage medium.

A computer-readable storage medium, wherein a computer program is stored in the computer-readable storage medium, and when the computer program is loaded and executed by a processor, the above-mentioned comprehensive risk warning method for mountain torrents is adopted.

Among them, the computer program can be stored in a computer-readable medium, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file or some middleware, etc. Any entity or device carrying computer program code, recording medium, USB flash drive, removable hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random-access memory (RAM), electrical carrier signal, telecommunication signal and software It should be noted that the computer-readable medium includes but is not limited to the above components.

Wherein, through the computer-readable storage medium, the mountain torrent comprehensive risk early warning method in the above embodiment is stored in the computer-readable storage medium, and loaded and executed on the processor, so as to facilitate the storage and application of the above-mentioned method.

All of the above are preferred embodiments of the present application, and are not intended to limit the protection scope of the application. Therefore, all equivalent changes made according to the structure, shape and principle of the application should be covered by the protection scope of the application. Inside.

Claims (9)

1. A mountain torrent comprehensive risk early warning method is characterized by comprising the following steps:

acquiring attribute information of a target small watershed based on a preset torrential flood risk basic attribute data set;

classifying the attribute information to obtain attribute factors; the attribute factors comprise rainfall factors, terrain factors and social factors;

constructing a correlation coefficient matrix for the attribute factors;

performing principal component analysis on the attribute factors based on the correlation coefficient matrix to obtain a plurality of key factors;

normalizing the plurality of key factors to obtain the weight of the risk factor;

calculating a risk index based on the risk factor weight;

and determining the early warning grade corresponding to the risk index based on a preset risk index range.

2. The method for integrated risk early warning of torrential floods according to claim 1, wherein the torrential flood risk basic attribute data set comprises torrential rain atlas vector data, small watershed attribute data and forecast rainfall data;

before the classifying the attribute information, the method includes:

combining the stormwater atlas vector data with the small watershed attribute data;

and interpolating the forecast rainfall data into grid data with preset resolution based on an inverse distance weight method.

3. The method according to claim 1, wherein the performing principal component analysis on the attribute factors based on the correlation coefficient matrix to obtain a plurality of key factors comprises:

calculating the factor sensitivity of the attribute factors based on a preset principal component analysis model and a preset certainty coefficient model;

determining key factors according to the factor sensitivity.

4. The method for comprehensive risk early warning of torrential floods according to claim 3, wherein the calculating the factor sensitivity of the attribute factor based on a preset principal component analysis model and a preset certainty coefficient model comprises:

calculating a first ratio of the number of disasters causing the mountain torrent disasters to the preset area of the target area under each attribute factor;

calculating a second ratio of the number of the disasters under each attribute factor to a preset area of a hill area;

respectively calculating the certainty factors of the rainfall factor, the terrain factor and the social factor based on a first ratio, a second ratio and a preset certainty factor model formula;

and calculating the factor sensitivity of the attribute factor based on a preset principal component analysis model and the certainty coefficient.

5. The mountain torrent comprehensive risk early warning method according to claim 4, wherein the method comprises the following steps: the deterministic coefficient model formula is:

wherein PPa is the first ratio; PPs is the second ratio; CF is a deterministic coefficient.

6. The method for mountain torrent comprehensive risk early warning according to claim 1, wherein the calculating a risk index based on the risk factor weight comprises:

calculating the weight of the risk factor based on an objective and subjective combination method;

and adding the weights of the risk factors to obtain a risk index.

7. The method as claimed in claim 1, wherein after determining the warning level corresponding to the risk index within the preset warning level range, the method comprises:

and marking a target area corresponding to the risk early warning in a preset area map.

8. The utility model provides a risk early warning system is synthesized to torrential flood which characterized in that includes:

the acquisition module is used for acquiring the attribute information of the target small watershed based on a preset torrent risk basic attribute data set;

the classification module is used for classifying the attribute information to obtain an attribute factor;

a construction module for constructing a correlation coefficient matrix for the attribute factors;

the analysis module is used for carrying out principal component analysis on the attribute factors based on the correlation coefficient matrix to obtain a plurality of key factors;

the processing module is used for carrying out normalization processing on the plurality of key factors to obtain risk factor indexes;

a calculation module for calculating a risk index based on the risk factor indicator;

the determining module is used for determining the early warning grade corresponding to the risk index based on a preset risk index range.

9. A computer-readable storage medium, in which a computer program is stored, which, when loaded and executed by a processor, carries out the method of any one of claims 1 to 7.

Images