Forest smoke and fire alarm risk index calculation and risk assessment method based on multiple factors

Abstract

The invention discloses a forest smoke and fire alarm risk index calculation and risk assessment method based on multiple factors. The method comprises the steps of obtaining monitoring data of various meteorological factors, measurement data of various geographical environment factors and detection data of various human activity factors according to position information of an alarm point; and calculating a meteorological risk index of smoke and fire alarm based on the monitoring data and the corresponding fitting regression coefficient. And correcting the meteorological risk index through the rainfall correction coefficient and the snowfall correction coefficient. And correction coefficients corresponding to the geological environment factors and the human activity factors are set according to the measured data. In combination with correction coefficients of all geological environment factors and human activity factors, the corrected meteorological risk index is further corrected to obtain a comprehensive risk index of smoke and fire alarm, so that the accuracy of a quantitative evaluation result of the smoke and fire alarm risk is improved, and the result is used as a basis for alarm verification of forest protection personnel; and thus, the alarm verification workload of forest rangers is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of fire alarms, and particularly relates to a method for calculating a forest fire and smoke alarm risk index based on multiple factors and risk assessment. Background Art

[0002] As an important disturbance factor in forest ecosystems, forest fires will seriously damage the structure and function of forest ecosystems. Timely detection of fire and carrying out fire extinguishing work is of great significance for forest protection. In recent years, with the popularization and application of Internet of Things devices such as forestry high-altitude cloud platforms and high-definition video monitoring, the intelligent recognition and positioning algorithm of fire and smoke based on video images has also developed rapidly, and basically can replace the manual method to quickly detect fire.

[0003] However, due to more cloudy and foggy weather in some mountainous areas, and the interference of production and living fires such as burning ash for fertilizer accumulation and cooking at the forest boundary, the daily number of intelligent alarms is relatively large. It still requires forest rangers to manually verify and determine one by one, resulting in a large pressure on verification work and a reduction in management efficiency. Summary of the Invention

[0004] Aiming at the deficiencies of the existing technology, the present invention proposes a method for calculating a forest fire and smoke alarm risk index based on multiple factors and risk assessment, which can reduce the burden of alarm verification work for forest rangers and improve the forestry management efficiency. The specific technical solutions are as follows:

[0005] In the first aspect, a method for calculating a forest fire and smoke alarm risk index based on multiple factors is provided. In the first feasible implementation manner of the first aspect, it includes:

[0006] Obtain the monitoring data of various meteorological factors, the measurement data of various geological environment factors, and the detection data of various human activity factors at the alarm point according to the location information of the alarm point in the fire and smoke alarm information;

[0007] Calculate the meteorological risk index of the fire and smoke alarm information through the monitoring data of the corresponding meteorological factors and their corresponding fitting regression coefficients;

[0008] Set a rainfall correction coefficient and a snowfall correction coefficient, and correct the meteorological risk index according to the rainfall correction coefficient and the snowfall correction coefficient;

[0009] Set the correction coefficients corresponding to the various geological environment factors according to the corresponding measurement data, and set the correction coefficients corresponding to the various human activity factors according to the corresponding detection data;

[0010] Combine the correction coefficients corresponding to all the geological environment factors and human activity factors to further correct the corrected meteorological risk index to obtain the comprehensive risk index of the fire and smoke alarm information.

[0011] Combined with the first implementation manner of the first aspect, in the second implementation manner of the first aspect, obtaining the monitoring data, measurement data, and detection data at the alarm point includes:

[0012] Adopt an intelligent recognition and positioning algorithm for fireworks based on video images to identify and locate the alarm points in the monitoring area, and determine the position information of the alarm points.

[0013] Combined with the first implementation manner of the first aspect, in the third implementation manner of the first aspect, obtaining the monitoring data of various meteorological factors at the alarm point includes:

[0014] Obtain the real-time monitoring information of each meteorological monitoring station around the alarm point according to the position information;

[0015] Based on the real-time monitoring information of each meteorological monitoring station, use the spatial interpolation algorithm to determine the monitoring data of various meteorological factors at the alarm point.

[0016] Combined with the first implementation manner of the first aspect, in the fourth implementation manner of the first aspect, obtaining the measurement data of various geological environment factors at the alarm point includes:

[0017] Obtain the land use type and digital elevation model at the alarm point according to the position information;

[0018] Based on the land use type and digital elevation model, use the GIS spatial interpolation method, spatial address analysis method, or nearest neighbor analysis method to determine the measurement data of various geological environment factors at the alarm point.

[0019] Combined with the first implementation manner of the first aspect, in the fifth implementation manner of the first aspect, calculating the meteorological risk index of the fireworks alarm information includes:

[0020] Collect the historical monitoring information of multiple alarm points where historical fires occurred, as well as the historical monitoring information of multiple alarm points where no fires occurred, and construct a data set;

[0021] Train a regression analysis model with the fire point data as the dependent variable and the meteorological data as the independent variable through the constructed data set to obtain a trained fire occurrence probability model;

[0022] Determine the fitting regression coefficients corresponding to various meteorological factors through the fire occurrence probability model.

[0023] Combined with the fifth implementation manner of the first aspect, in the sixth implementation manner of the first aspect, adopt the Logistic model as the regression analysis model.

[0024] Combined with the first implementation manner of the first aspect, in the seventh implementation manner of the first aspect, setting a rainfall correction coefficient and a snowfall correction coefficient includes:

[0025] Respectively compare the monitoring data of rainfall and snowfall at the alarm point with the corresponding thresholds, and set the rainfall correction coefficient and the snowfall correction coefficient according to the comparison results.

[0026] Combined with the first implementation manner of the first aspect, in the eighth implementation manner of the first aspect, setting the correction coefficient corresponding to the geological environment factor through the corresponding measurement data includes:

[0027] Set the correction coefficients corresponding to the geographical environment factor in different evaluation intervals, and different evaluation intervals respectively correspond to different measurement data ranges;

[0028] Match the measurement data with the measurement data ranges corresponding to each evaluation interval, and set the correction coefficient corresponding to the geological environment factor according to the matching result.

[0029] Combined with the eighth implementation manner of the first aspect, in the ninth implementation manner of the first aspect, setting the correction coefficients corresponding to the geographical environment factor in different evaluation intervals includes:

[0030] Statistically analyze the historical measurement data of the geographical environment factors of multiple alarm points where fires have occurred to determine the factor variation range corresponding to the geographical environment factor;

[0031] Equally divide the factor variation range to determine the measurement data ranges corresponding to each evaluation interval of the geographical environment factor;

[0032] Count the occurrence times of fires and the occurrence times of geographical environment factors in each evaluation interval according to the measurement data ranges, and calculate the risk levels of each evaluation interval according to the corresponding occurrence times and appearance times;

[0033] Normalize the risk levels of all evaluation intervals to obtain the correction coefficients corresponding to the geographical environment factor in different evaluation intervals.

[0034] In a second aspect, a forest fire and smoke alarm risk assessment method based on multiple factors is provided, including:

[0035] Adopt the forest fire and smoke alarm risk index calculation method described in any one of the first to ninth implementation manners of the first aspect to calculate the comprehensive risk index of the fire and smoke alarm information;

[0036] Quantitatively evaluate the risk level of the fire and smoke alarm according to the comprehensive risk index.

[0037] Beneficial effects: By adopting the multi-factor-based forest fire alarm risk index calculation and risk assessment method of the present invention, it is possible to quantitatively evaluate the fire alarm risk at the alarm point identified and located by the intelligent fire and smoke identification and positioning algorithm based on the real-time monitoring data of various meteorological factors at the alarm point, and comprehensively correct the quantitative evaluation result of the fire risk at the alarm point through the rainfall and snowfall conditions and various geographical environment factors and human activity factors at the alarm point, so as to improve the accuracy of the quantitative evaluation result of the fire alarm risk, serve as the basis for forest rangers to verify alarms, and further reduce the work burden of forest rangers in verifying alarms and improve the efficiency of forestry management. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the specific embodiments of the present invention, the drawings required for the specific embodiments will be briefly introduced below. In all the drawings, the components or parts are not necessarily drawn to scale.

[0039] Figure 1 It is a flowchart of a multi-factor-based forest fire alarm risk index calculation method provided by an embodiment of the present invention;

[0040] Figure 2 It is a flowchart of setting the correction coefficients for each evaluation interval provided by an embodiment of the present invention;

[0041] Figure 3 It is a flowchart of a multi-factor-based forest fire alarm risk assessment method provided by an embodiment of the present invention. SPECIFIC EMBODIMENTS

[0042] The embodiments of the technical solution of the present invention will be described in detail below with reference to the drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, so they are only examples and cannot be used to limit the protection scope of the present invention.

[0043] It should be understood that meteorological factors are important factors for the occurrence of forest fires. Currently, most mainstream forest fire danger level warning models are calculated by the meteorological department based on large-scale meteorological data collected by meteorological satellites. Therefore, in this embodiment, various meteorological factors can be used as the main factors to quantitatively evaluate the fire risk at the alarm point.

[0044] It should also be understood that the geological environment also has a great impact on the occurrence of fires. As the altitude increases, the precipitation increases, the temperature decreases, the evaporation decreases, the humidity increases, and the possibility of forest fires decreases. The influence of slope on the occurrence and spread of forest fires is basically high in the middle and low on both sides. As the slope increases, the surface runoff is fast, the combustibles on the ground are easily dried, the possibility of fire occurrence is greater, and the danger level is higher. However, when the slope reaches a steep slope (slope greater than 35°) or above, the mortality rate of forest trees caused by forest fires decreases, and the danger level decreases.

[0045] It should also be understood that human activities are also important causes leading to forest fires, and most fires are caused by humans. The risk of forest fires is higher in forest areas closer to settlements and roads. Therefore, in this embodiment, the quantitative assessment results of the fire risk at the alarm point can be corrected with various geological environment factors and human activity factors to improve the accuracy of the quantitative assessment results of the smoke and fire alarm risk.

[0046] Such as Figure 1 The flowchart of the calculation method for the forest smoke and fire alarm risk index based on multiple factors shown in the figure, and the calculation method includes:

[0047] Step 1: Obtain the monitoring data of various meteorological factors, the measurement data of various geological environment factors, and the detection data of various human activity factors at the alarm point according to the location information of the alarm point in the smoke and fire alarm information;

[0048] Step 2: Calculate the meteorological risk index of the smoke and fire alarm information through the monitoring data of the corresponding meteorological factors and their corresponding fitting regression coefficients;

[0049] Step 3: Set the rainfall correction coefficient and the snowfall correction coefficient, and correct the meteorological risk index according to the rainfall correction coefficient and the snowfall correction coefficient;

[0050] Step 4: Set the correction coefficients corresponding to the various geological environment factors according to the corresponding measurement data, and set the correction coefficients corresponding to the various human activity factors according to the corresponding detection data;

[0051] Step 5: Combine the correction coefficients corresponding to all the geological environment factors and human activity factors to further correct the corrected meteorological risk index to obtain the comprehensive risk index of the smoke and fire alarm information.

[0052] Specifically, first, the monitoring data of various meteorological factors, the measurement data corresponding to various geographical environment factors at the alarm point, and the detection data corresponding to various human activity factors can be collected in real time according to the location information of the alarm point in the smoke and fire alarm information. Then, the meteorological risk index of the fire occurrence at the alarm point can be calculated based on the monitoring data of the various meteorological factors and their corresponding fitting regression coefficients to quantitatively evaluate the smoke and fire alarm risk at the alarm point identified by the intelligent smoke and fire recognition and positioning algorithm.

[0053] After that, the calculated meteorological risk index can be corrected by the set rainfall correction coefficient and snowfall correction coefficient. Then, the correction coefficients corresponding to each of the geological environment factors at the alarm point can be set according to the corresponding measurement data, and the correction coefficients corresponding to each of the human activity factors can be set according to the corresponding detection data. Finally, combining the correction coefficients corresponding to all the geological environment factors and human activity factors, the corrected meteorological risk index is further corrected to obtain the comprehensive risk index of the smoke and fire alarm information, thereby improving the accuracy of the quantitative assessment result of the smoke and fire alarm risk, serving as the basis for forest rangers to verify alarms, and further reducing the workload of forest rangers for alarm verification and enhancing the efficiency of forestry management.

[0054] In this embodiment, comprehensively considering the impacts of meteorological conditions, geological environment, and human activities on forest fires, 11 factors such as air temperature and relative humidity are selected as the basis for smoke and fire risk assessment, as shown in the following table:

[0055]

[0056] It should be understood that only these 11 factors are used for illustration in this embodiment, but the present invention is not limited thereto, and other factors can also be selected or added as the basis for smoke and fire risk assessment.

[0057] In this embodiment, optionally, in step 1, the monitoring data, measurement data, and detection data at the alarm point are obtained, including:

[0058] Using a smoke and fire intelligent recognition and positioning algorithm based on video images, the alarm point within the monitoring area is recognized and positioned to determine the location information of the alarm point.

[0059] Specifically, when obtaining the data of each factor at the alarm point according to the location information of the alarm point, the video images of the monitoring area can be collected in real time, and a smoke and fire intelligent recognition and positioning algorithm is used to locate and identify the suspected ignition point in the video image, that is, the location of the alarm point. In this embodiment, an existing smoke and fire intelligent recognition and positioning algorithm is adopted, and the specific step principle thereof belongs to the common knowledge in the art and will not be elaborated here.

[0060] In this embodiment, optionally, in step 1, the monitoring data of each meteorological factor at the alarm point are obtained, including:

[0061] Obtaining the real-time monitoring information of each meteorological monitoring station around the alarm point according to the location information;

[0062] Based on the real-time monitoring information of each meteorological monitoring station, a spatial interpolation algorithm is used to determine the monitoring data of each meteorological factor at the alarm point.

[0063] Specifically, when obtaining the monitoring data of various meteorological factors at the alarm point, according to the location information of the alarm point, the real-time monitoring data of various meteorological factors monitored by each meteorological monitoring station around the alarm point can be collected, and based on the real-time monitoring data of each meteorological monitoring station, the monitoring data corresponding to various meteorological factors can be calculated by using a spatial interpolation algorithm.

[0064] The spatial interpolation algorithm is a method of converting the measurement data of discrete points into a continuous data surface. By using the spatial interpolation algorithm, the monitoring data of various meteorological factors at the alarm point can be deduced from the real-time monitoring data of various meteorological factors measured by each meteorological monitoring station. Specifically, methods such as the moving average method and the inverse distance squared weighted method can be used to deduce the monitoring data of various meteorological factors at the alarm point. The specific steps belong to the common knowledge in this field and will not be elaborated here.

[0065] In this embodiment, optionally, in step 1, obtaining the measurement data of various geological environment factors at the alarm point includes:

[0066] Obtaining the land use type and digital elevation model at the alarm point according to the location information;

[0067] Based on the land use type and digital elevation model, using the GIS spatial interpolation method, spatial address analysis method or nearest neighbor analysis method, determining the measurement data of various geological environment factors at the alarm point.

[0068] Specifically, when obtaining the measurement data of various geological environment factors at the alarm point, the 30m resolution land use data and 30m resolution digital elevation model at the alarm point can be retrieved according to the location information first. Among them, the land use data includes the land use type, and the digital elevation model is a digital simulation of the ground terrain through limited terrain elevation data, including various geomorphic factors such as elevation, such as the spatial distribution of linear and non-linear combinations of factors such as slope, aspect, and slope change rate.

[0069] Based on the digital elevation model, using the GIS spatial interpolation method, spatial address analysis method or nearest neighbor analysis method, calculating the measurement data of geological environment factors such as slope and altitude at the alarm point, and the distance between the alarm point and the residential area and the distance between the alarm point and the road can be calculated.

[0070] In this embodiment, optionally, in step 2, calculating the meteorological risk index of the smoke alarm information includes:

[0071] Collecting the historical monitoring information of multiple alarm points where historical fires occurred and the historical monitoring information of multiple alarm points where no fires occurred, and constructing a data set;

[0072] Training a regression analysis model with the fire point data as the dependent variable and the meteorological data as the independent variable through the constructed data set to obtain a trained fire occurrence probability model;

[0073] Determining the fitting regression coefficients corresponding to each meteorological factor through the fire occurrence probability model.

[0074] Specifically, by performing a regression analysis on the monitoring data of each meteorological factor at a large number of alarm points where fires occurred and alarm points where no fires occurred, the correlation between each meteorological factor and the fire risk can be determined to quantitatively evaluate the influence degree of the meteorological factor at the alarm point on the fire risk, that is, the fitting regression coefficient corresponding to the meteorological factor.

[0075] Combining the monitoring data of each meteorological factor and the fitting regression coefficients, the meteorological risk index at the alarm point can be calculated. The specific calculation formula is as follows:

[0076] U = β V V + β T T + β RH r RH + β M M;

[0077] Wherein, V, T, r RH , M are the real-time wind speed, real-time temperature, real-time relative humidity and the number of consecutive days without precipitation at the alarm point respectively. β V , β T , β RH , β M are the fitting regression coefficients corresponding to the real-time wind speed, real-time temperature, real-time relative humidity and the number of consecutive days without precipitation respectively.

[0078] In this embodiment, the fitting regression coefficients corresponding to the above-mentioned meteorological factors can be calculated by using a regression analysis model based on historical fire information. Specifically, taking the fire point data as the dependent variable of the regression model and the meteorological data as the independent variable, a part of the alarm points without fires is selected as random points at a ratio of 1:1 and mixed with the historical real fire points as the data set. The modeling part and the verification part are segmented from the data set. Training the regression analysis model through the modeling part, and then testing the training of the regression analysis model obtained through the verification part. Finally, the calculation model of the fire occurrence probability model under multiple meteorological factors is:

[0079]

[0080] Wherein, β 0 is the regression constant, and the fitting regression coefficients corresponding to each meteorological factor can be determined through the fire occurrence probability model.

[0081] In this embodiment, optionally, the Logistic model is used as the regression analysis model. Specifically, the regression analysis model includes a linear regression model, a multiple regression model, and a Logistic model. In this embodiment, the determination result has only two states, "fire" and "non-fire", which is a typical binary variable, and logistic regression is often used to study the relationship between binary, multi-class, and ordered multi-class observation results (dependent variables) and some influencing factors (independent variables). Therefore, the Logistic model can be selected to construct the fire occurrence probability model.

[0082] In this embodiment, optionally, in step 3, the rainfall correction coefficient and the snowfall correction coefficient are set, including:

[0083] The monitoring data of rainfall and snowfall at the alarm point are respectively compared with the corresponding thresholds, and the rainfall correction coefficient and the snowfall correction coefficient are set according to the comparison results.

[0084] Specifically, considering that the probability of forest fire occurrence is relatively low under precipitation and snowfall conditions, the rainfall correction coefficient and the snowfall correction coefficient are added to the meteorological risk index for correction to improve the accuracy of the quantitative evaluation result of the smoke and fire alarm. The correction calculation formula is as follows:

[0085] I = U × C r × C s ;

[0086] Where C r is the rainfall correction coefficient, and C s is the snowfall correction coefficient.

[0087] The rainfall correction coefficient and the snowfall correction coefficient are related to the rainfall amount and the snow accumulation amount at the alarm point. When setting the rainfall correction coefficient and the snowfall correction coefficient, the rainfall amount and the snow accumulation amount at the alarm point can be respectively compared with the set thresholds. When the rainfall amount R r ≥ 2 mm per hour at the alarm point, the rainfall correction coefficient C r = 0, when R r < 2 mm, C r = 1. When the depth H s > 0 mm of snow accumulation per hour at the alarm point, the snowfall correction coefficient C s = 0, when the depth of snow accumulation H s = 0 mm, the snowfall correction coefficient C s = 1.

[0088] In this embodiment, optionally, in step 3, the correction coefficient corresponding to the geological environment factor is set through the corresponding measurement data, including:

[0089] Set the correction coefficients corresponding to different evaluation intervals of the geographical environment factor, where different evaluation intervals correspond to different measurement data ranges;

[0090] Match the measurement data with the measurement data ranges corresponding to each evaluation interval, and set the correction coefficient corresponding to the geological environment factor according to the matching result.

[0091] Specifically, when setting the correction coefficient of the geological environment factor, the measurement data of the geological environment factor at the alarm point can be matched with the measurement data ranges corresponding to different correction coefficient evaluation intervals set in advance, so as to determine the evaluation interval to which the measurement data of the geological environment factor at the alarm point belongs, and further determine the correction coefficient corresponding to the geological environment factor at the alarm point.

[0092] In this embodiment, optionally, setting the correction coefficients corresponding to different evaluation intervals of the geographical environment factor includes:

[0093] Step 3-1: Statistically analyze the historical measurement data of the geographical environment factor at multiple alarm points where fires have occurred to determine the factor variation range corresponding to the geographical environment factor;

[0094] Step 3-2: Perform equal-interval segmentation on the factor variation range to determine the measurement data ranges corresponding to each evaluation interval of the geographical environment factor;

[0095] Step 3-3: Count the number of fire occurrences and the number of occurrences of the geographical environment factor in each evaluation interval according to the measurement data range, and calculate the risk level of each evaluation interval according to the corresponding number of occurrences and the number of appearances;

[0096] Step 3-4: Normalize the risk levels of all evaluation intervals to obtain the correction coefficients corresponding to different evaluation intervals of the geographical environment factor.

[0097] Specifically, as Figure 2 shown, when setting the correction coefficients corresponding to different evaluation intervals of the geological environment factor, the historical measurement data of the geological environment factor at multiple alarm points where fires have occurred can be collected first, and all the obtained historical measurement data can be statistically analyzed to determine the maximum value and minimum value of the geological environment factor, and then the factor variation range of the geological environment factor can be determined according to the maximum value and minimum value. Then, the factor variation range of the geological environment factor can be equally segmented into multiple different evaluation intervals, and each evaluation interval corresponds to a different measurement data range. After that, the risk level of fires occurring in different evaluation intervals of the geological environment factor can be calculated. The specific calculation formula is as follows:

[0098] P i =F i / Yi ;

[0099] Among them, F i is the number of fire occurrences within the evaluation interval, Y i is the number of occurrences of geological environment factors within the evaluation interval, and i is the number of evaluation intervals.

[0100] Finally, normalize the risk levels of each evaluation interval to obtain the relative risks corresponding to the geological environment factors within different measurement data ranges, that is, the correction coefficients corresponding to the geological environment factors in different evaluation intervals. The specific calculation formula is as follows:

[0101] C i = P i / P max ;

[0102] Among them, P max is the maximum value of the risk levels corresponding to all evaluation intervals of the geological environment factor.

[0103] In this embodiment, the same setting method as the correction coefficient of the geographical environment factor in different evaluation intervals can be adopted to set the correction coefficient of the human activity factor in different evaluation intervals, match the detection data of the human activity factor with the detection data ranges corresponding to different evaluation intervals, and set the correction coefficients of each human activity factor at the alarm point according to the matching results.

[0104] After the correction coefficients of each geological environment factor and human activity factor at the alarm point are set, based on the correction coefficients corresponding to all influencing factors, further correct the corrected meteorological risk index, and finally obtain the comprehensive risk index of the smoke and fire alarm information, further improving the accuracy of the quantitative evaluation result of the smoke and fire alarm. The specific technical solution is as follows:

[0105] I c = I × C a × C b × C c × C d × C e × C f ;

[0106] Among them, C a , C b , C c , C d , C e , C f are the correction coefficients corresponding to slope, altitude, land use type, distance to residential area, distance to road, and holiday respectively.

[0107] The calculated comprehensive risk index can be used as the basis for forest rangers to verify alarms, thereby reducing the workload of forest rangers in verifying alarms and improving the efficiency of forestry management.

[0108] Such as Figure 3 The flowchart of the multi-factor-based forest fire and smoke alarm risk assessment method shown, the assessment method includes:

[0109] Step S1: Use the above-mentioned forest fire and smoke alarm risk index calculation method to calculate the comprehensive risk index of the fire and smoke alarm information;

[0110] Step S2: Quantitatively evaluate the risk level of the fire and smoke alarm according to the comprehensive risk index.

[0111] Specifically, first, the multi-factor-based forest fire and smoke alarm risk index calculation method of this embodiment can be used to calculate the comprehensive risk index of the fire and smoke alarm. Then, the calculated comprehensive risk index can be matched with the risk index range corresponding to each fire risk level, and the risk level of the fire and smoke alarm can be determined according to the matching result.

[0112] In this way, a quantitative evaluation can be carried out on the fire and smoke alarms generated by the intelligent fire and smoke identification and positioning algorithm. The risk index and risk level of the fire and smoke alarm are calculated by comprehensively using meteorological conditions, geological environment, and human activity-related factors, and used as the basis for forest rangers to verify, reducing the workload of forest rangers in verifying alarms and improving the efficiency of forestry management. The risk index ranges corresponding to different risk levels are shown in the following table:

[0113]

[0114] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be covered by the scope of the claims and the description of the present invention.

Claims

1. A method for calculating forest fire alarm risk index based on multiple factors, characterized in that: include: Acquire monitoring data of various meteorological factors, measurement data of various geological environmental factors, and detection data of various human activity factors at the alarm point according to the location information of the alarm point in the fireworks alarm information; Calculate the meteorological risk index of the fireworks alarm information through the monitoring data of the corresponding meteorological factors and the corresponding fitting regression coefficients; Setting a rainfall correction coefficient and a snow correction coefficient, and correcting the meteorological risk index according to the rainfall correction coefficient and the snow correction coefficient; The correction coefficients corresponding to the various geological environmental factors are set according to the corresponding measurement data, and the correction coefficients corresponding to the various human activity factors are set according to the corresponding detection data; The revised meteorological risk index is further corrected by combining the correction coefficients corresponding to all the geological environmental factors and human activity factors to obtain a comprehensive risk index for the fireworks alarm information.

2. The method for calculating the forest fire alarm risk index based on multiple factors according to claim 1 is characterized in that: Obtain monitoring data, measurement data and detection data at the alarm point, including: A firework intelligent recognition and positioning algorithm based on video images is used to identify and locate alarm points within the monitoring area and determine the location information of the alarm points.

3. The method for calculating forest fire alarm risk index based on multiple factors according to claim 1 is characterized in that: Obtain monitoring data of various meteorological factors at the alarm point, including: Acquire real-time monitoring information of each meteorological monitoring station around the alarm point according to the location information; Based on the real-time monitoring information of each meteorological monitoring station, a spatial interpolation algorithm is used to determine the monitoring data of each meteorological factor at the alarm point.

4. The method for calculating forest fire alarm risk index based on multiple factors according to claim 1, characterized in that: Obtain measurement data of various geological environmental factors at the alarm point, including: Acquire the land use type and digital elevation model at the alarm point according to the location information; Based on the land use type and digital elevation model, GIS spatial interpolation method, spatial address analysis method or nearest neighbor analysis method is used to determine the measurement data of various geological environmental factors at the alarm point.

5. The method for calculating forest fire alarm risk index based on multiple factors according to claim 1 is characterized in that: Calculating the meteorological risk index of the fire alarm information includes: Collect historical monitoring information of multiple alarm points where historical fires occurred, and historical monitoring information of multiple alarm points where no fires occurred, to build a data set; Through the constructed data set, the regression analysis model with fire point data as dependent variable and meteorological data as independent variable is trained to obtain the trained fire occurrence probability model; The fitting regression coefficients corresponding to various meteorological factors are determined through the fire occurrence probability model.

6. The method for calculating the forest fire alarm risk index based on multiple factors according to claim 5 is characterized in that: The Logistic model was used as the regression analysis model.

7. The method for calculating forest fire alarm risk index based on multiple factors according to claim 1, characterized in that: Set the rain correction factor and snow correction factor, including: The monitoring data of rainfall and snowfall at the alarm point are respectively compared with corresponding threshold values, and the rainfall correction coefficient and snowfall correction coefficient are set according to the comparison results.

8. The method for calculating forest fire alarm risk index based on multiple factors according to claim 1, characterized in that: The correction coefficients corresponding to the geological environment factors are set through the corresponding measurement data, including: Setting correction coefficients corresponding to the geographical environment factors in different evaluation intervals, where different evaluation intervals correspond to different measurement data ranges; The measurement data is matched with the measurement data range corresponding to each evaluation interval, and the correction coefficient corresponding to the geological environment factor is set according to the matching result.

9. The method for calculating forest fire alarm risk index based on multiple factors according to claim 8, characterized in that: The correction coefficients corresponding to the geographical environment factors in different evaluation intervals are set, including: Performing statistical analysis on historical measurement data of geographical environmental factors at multiple alarm points where fires have occurred, and determining factor variation ranges corresponding to the geographical environmental factors; The factor variation range is divided into equal intervals to determine the measurement data range corresponding to each evaluation interval of the geographical environment factor; Counting the number of fire occurrences and the number of occurrences of geographical environmental factors in each evaluation interval according to the measurement data range, and calculating the danger level of each evaluation interval according to the corresponding number of occurrences and numbers of occurrences; The danger levels of all evaluation intervals are normalized to obtain correction coefficients corresponding to the geographical environment factors in different evaluation intervals.

10. A multi-factor-based forest fire alarm risk assessment method, characterized in that: include: Using the forest fire alarm risk index calculation method according to any one of claims 1 to 9 to calculate the comprehensive risk index of the fire alarm information; The risk level of the smoke and fire alarm is quantitatively evaluated according to the comprehensive risk index.

Images