Forest fire comprehensive risk prediction method and device and electronic equipment

Abstract

The invention provides a forest fire comprehensive risk prediction method and apparatus, and an electronic device. The method comprises the steps of determining a value of a precipitation amount correction coefficient and a value of an accumulated snow amount correction coefficient, and determining whether fire point information is monitored; under the condition that the fire point information is not monitored, obtaining prediction data of each index factor in a forest comprehensive fire danger evaluation index system of the to-be-evaluated forest region; for the prediction data of each index factor, performing quantization processing on the prediction data based on the quantization rule of the index factor to obtain an index quantization value of the index factor; inputting the index quantized value of each index factor, the value of the precipitation amount correction coefficient and the value of the snow accumulation amount correction coefficient in the forest comprehensive fire danger evaluation index system into a forest comprehensive fire danger prediction model to obtain a forest comprehensive fire danger value of the forest region to be evaluated; and obtaining a forest comprehensive fire danger grade of the forest region to be evaluated based on the forest comprehensive fire danger value and a forest fire risk grade division threshold. And the accuracy of forest fire comprehensive risk prediction is improved.

Description

Technical Field

[0001] The present application relates to the technical field of forest fire risk assessment and prediction, and in particular to a method, device and electronic equipment for comprehensive forest fire risk prediction. Background Art

[0002] Forest fire risk warning is the forerunner of forest fire prevention and control. Foreign forest fire risk prediction research began in the 1930s. Currently, the forest fire risk forecasting models that have been put into commercial use are mostly based on meteorological data provided by meteorological departments, such as the Canadian Forest Service's forest fire weather index system, Australia's forest fire risk rating system, the United States' national forest fire risk rating system, and Russia's forest fire risk prediction system. Domestic forest fire risk prediction research is mainly based on forest fire risk research in Europe and the United States, and is combined with domestic actual conditions to carry out in-depth research, such as the "comprehensive index method", "801 fire risk scale method" and "forest fire risk meteorological level".

[0003] However, the existing comprehensive forest fire risk prediction methods are not comprehensive enough in terms of consideration and prediction accuracy, and need to be improved. Summary of the invention

[0004] The present application aims to solve one of the technical problems in the related art at least to some extent.

[0005] To this end, the first purpose of this application is to propose a forest fire comprehensive risk prediction method to improve the accuracy of forest fire comprehensive risk prediction.

[0006] The second purpose of this application is to provide a comprehensive forest fire risk prediction device.

[0007] The third objective of the present application is to provide an electronic device.

[0008] A fourth objective of the present application is to provide a computer-readable storage medium.

[0009] A fifth object of the present application is to provide a computer program product.

[0010] To achieve the above objectives, the first embodiment of the present application proposes a forest fire comprehensive risk prediction method, including:

[0011] Determine the value of the precipitation correction factor and the value of the snow accumulation correction factor based on the precipitation data and the snow accumulation data of the forest area to be evaluated in the first time period;

[0012] Determining whether fire point information is monitored in the forest area to be assessed;

[0013] In the case that no fire point information is monitored in the forest area to be evaluated, the prediction data of each indicator factor in the forest comprehensive fire risk assessment index system of the forest area to be evaluated in the second time period in the future is obtained; and for the prediction data of each indicator factor, the prediction data is quantified based on the quantization rule of the indicator factor to obtain the indicator quantization value of the indicator factor;

[0014] Inputting the index quantification value of each index factor in the comprehensive forest fire risk assessment index system, the value of the precipitation correction coefficient and the value of the snow accumulation correction coefficient into the comprehensive forest fire risk prediction model to obtain the comprehensive forest fire risk value of the forest area to be assessed;

[0015] Based on the comprehensive forest fire risk value of the forest area to be evaluated and the forest fire risk level classification threshold, the comprehensive forest fire risk level of the forest area to be evaluated is obtained.

[0016] In some embodiments, the primary index of the comprehensive forest fire risk assessment index system includes a hazard index and an exposure index.

[0017] The secondary indicators of the hazard index include vegetation factors, meteorological factors, terrain factors and fire source factors. The tertiary indicators of the vegetation factors include vegetation type and EDVI double-layer vegetation index. The tertiary indicators of the meteorological factors include precipitation, snow accumulation, wind speed, temperature, relative humidity and drought level. The tertiary indicators of the terrain factors include altitude, slope and slope direction. The tertiary indicators of the fire source factors include natural fire sources and man-made fire sources. The quaternary indicators of man-made fire sources include the spatial distribution of settlements and road network density.

[0018] The secondary indicators of the exposure index include population density and GDP spatial distribution.

[0019] In some embodiments, the comprehensive forest fire risk prediction model is expressed as follows:

[0020]

[0021] Among them, P 综合火险 is the comprehensive forest fire risk value, C r is the precipitation correction coefficient. When the 24h precipitation is greater than or equal to 1mm, the value is 0; when the 24h precipitation is less than 1mm, the value is 1; C s is the snow depth correction coefficient. When the 24-hour snow depth is greater than 0, the value is 0; when the 24-hour snow depth is equal to 0, the value is 1; U FFDI W is the forest fire weather index; FFDI is the index weight of meteorological factors; A iis the quantitative value of the ith indicator, excluding meteorological factors, including 10 indicators: vegetation type, vegetation index, altitude, slope, slope aspect, natural fire source, spatial distribution of distance to residential areas, road network density, population density, and GDP spatial distribution; W i is the indicator weight of the ith indicator, excluding meteorological factors; n is the number of evaluation indicators, excluding meteorological factors; K is the fire point monitoring sensitivity. When a fire point is detected, the area where the fire point is located is directly considered to be extremely high risk, with a value of 100; if no fire point is detected, the value is 0;

[0022] The forest fire risk meteorological index is expressed as follows:

[0023] U FFDI =f(V)+f(T)+f(r RH )+f(M)

[0024] Where V is wind speed, T is temperature, r RH is the relative humidity, M is the number of consecutive days without precipitation, f(V) is the quantitative value of the wind speed index, f(T) is the quantitative value of the temperature index, and f(r RH ) is the quantitative value of the relative humidity index, and f(M) is the quantitative value of the continuous precipitation-free day index.

[0025] In some embodiments, the method for obtaining the indicator weight of each indicator factor in the comprehensive forest fire risk assessment indicator system includes:

[0026] Through the gray correlation method and the improved hierarchical analysis method, the index weights of each index factor in the forest fire risk assessment index system are obtained.

[0027] In some embodiments, the grayscale correlation method and the improved analytic hierarchy process are used to obtain the index weights of each index factor in the forest fire risk assessment index system. , including:

[0028] Obtain historical forest burnt area data and corresponding historical data of each indicator factor in the forest comprehensive fire risk assessment indicator system;

[0029] The historical forest burnt area data is used as a reference sequence, and the corresponding historical data of each indicator factor in the forest comprehensive fire risk assessment index system is used as a comparison sequence;

[0030] Based on the comparison sequence and the reference sequence, calculating the correlation coefficient and correlation degree of each indicator factor in the comprehensive forest fire risk assessment index system;

[0031] Based on the correlation coefficient and correlation degree of each indicator factor in the forest fire risk comprehensive assessment indicator system, the correlation order of the indicator factors in the forest fire risk comprehensive assessment indicator system is obtained; the correlation order represents the correlation degree ranking;

[0032] Based on the index level of each index factor in the forest fire risk comprehensive assessment index system, a judgment matrix is ​​constructed; the judgment matrix is ​​used to compare the importance of two index factors in the same index level; the importance of the index factor is obtained based on the correlation sequence of the index factors in the forest fire risk comprehensive assessment index system and a preset scale scoring standard; the preset scale scoring standard is a scale table of 1 to 20;

[0033] Obtaining the maximum eigenvalue of the judgment matrix, and calculating a random consistency index based on the maximum eigenvalue;

[0034] When the random consistency index meets the consistency test threshold, the square root method is used to calculate the maximum eigenvalue and eigenvector of the judgment matrix; and the weight of each indicator factor in the comprehensive forest fire risk assessment index system is determined based on the eigenvector.

[0035] In some embodiments, obtaining the comprehensive forest fire risk level of the forest area to be assessed based on the comprehensive forest fire risk value of the forest area to be assessed and the forest fire risk level classification threshold comprises:

[0036] If P 综合火险 When the fire risk is less than 39.8, the corresponding comprehensive forest fire risk level is level one, which is a low fire risk.

[0037] If 39.8≤P 综合火险 When the fire risk is less than 47.2, the corresponding comprehensive forest fire risk level is Level 2, which is a relatively low fire risk.

[0038] If 47.2≤P 综合火险 When the value is less than 66.4, the corresponding comprehensive forest fire risk level is level 3, which is a relatively high fire risk;

[0039] If 66.4≤P 综合火险 When the temperature is less than 73, the corresponding comprehensive forest fire risk level is level 4, which is a high fire risk;

[0040] If 73≤P 综合火险 When it is less than 100, the corresponding comprehensive forest fire risk level is level 5, which is an extremely high fire risk.

[0041] In some embodiments, the method further comprises:

[0042] In the case where fire point information is monitored in the forest area to be evaluated, determining a value of the fire point monitoring sensitivity;

[0043] The value of the fire point monitoring sensitivity, the value of the precipitation correction coefficient and the value of the snow accumulation correction coefficient are input into the comprehensive forest fire risk prediction model to obtain the comprehensive forest fire risk value of the forest area to be evaluated.

[0044] In some embodiments, the prediction data of the natural fire source in the future second time period includes lightning warning information; and the method for obtaining the index quantification value of the natural fire source includes:

[0045] Based on the lightning warning information and the lightning warning classification rules, a quantitative index value of the natural fire source is obtained;

[0046] The step of performing quantization processing on the predicted data based on the quantization rule of the indicator factor to obtain the indicator quantization value of the indicator factor includes:

[0047] Based on the classification rule of the indicator factor, determine the target category to which the predicted data of the indicator factor belongs;

[0048] Based on the category quantitative scoring rule of the indicator factor, the indicator quantitative value of the indicator factor is obtained.

[0049] To achieve the above-mentioned purpose, the second embodiment of the present application proposes a forest fire comprehensive risk prediction device, comprising:

[0050] A coefficient determination module, used to determine the value of the precipitation correction coefficient and the value of the snow accumulation correction coefficient based on the precipitation data and the snow accumulation data of the forest area to be evaluated in the first time period;

[0051] A fire point acquisition module is used to determine whether fire point information is monitored in the forest area to be evaluated;

[0052] An indicator acquisition module is used to obtain the forecast data of each indicator factor in the forest comprehensive fire risk assessment index system of the forest area to be assessed in the future second time period when no fire point information is monitored in the forest area to be assessed; and for the forecast data of each indicator factor, quantize the forecast data based on the quantization rule of the indicator factor to obtain the indicator quantization value of the indicator factor;

[0053] A fire risk assessment module is used to input the index quantification value of each index factor in the forest comprehensive fire risk assessment index system, the value of the precipitation correction coefficient and the value of the snow accumulation correction coefficient into the forest comprehensive fire risk prediction model to obtain the forest comprehensive fire risk value of the forest area to be assessed;

[0054] The level determination module is used to obtain the comprehensive forest fire risk level of the forest area to be evaluated based on the comprehensive forest fire risk value of the forest area to be evaluated and the forest fire risk level classification threshold.

[0055] In some embodiments, the primary index of the comprehensive forest fire risk assessment index system includes a hazard index and an exposure index.

[0056] The secondary indicators of the hazard index include vegetation factors, meteorological factors, terrain factors and fire source factors. The tertiary indicators of the vegetation factors include vegetation type and EDVI double-layer vegetation index. The tertiary indicators of the meteorological factors include precipitation, snow accumulation, wind speed, temperature, relative humidity and drought level. The tertiary indicators of the terrain factors include altitude, slope and slope direction. The tertiary indicators of the fire source factors include natural fire sources and man-made fire sources. The quaternary indicators of man-made fire sources include the spatial distribution of settlements and road network density.

[0057] The secondary indicators of the exposure index include population density and GDP spatial distribution.

[0058] In some embodiments, the comprehensive forest fire risk prediction model is expressed as follows:

[0059]

[0060] Among them, P 综合火险 is the comprehensive forest fire risk value, C r is the precipitation correction coefficient. When the 24h precipitation is greater than or equal to 1mm, the value is 0; when the 24h precipitation is less than 1mm, the value is 1; C s is the snow depth correction coefficient. When the 24-hour snow depth is greater than 0, the value is 0; when the 24-hour snow depth is equal to 0, the value is 1; U FFDI W is the forest fire weather index; FFDI is the index weight of meteorological factors; A i is the quantitative value of the ith indicator, excluding meteorological factors, including 10 indicators: vegetation type, vegetation index, altitude, slope, slope aspect, natural fire source, spatial distribution of distance to residential areas, road network density, population density, and GDP spatial distribution; W i is the indicator weight of the ith indicator, excluding meteorological factors; n is the number of evaluation indicators, excluding meteorological factors; K is the fire point monitoring sensitivity. When a fire point is detected, the area where the fire point is located is directly considered to be extremely high risk, with a value of 100; if no fire point is detected, the value is 0;

[0061] The forest fire risk meteorological index is expressed as follows:

[0062] U FFDI =f(V)+f(T)+f(r RH )+f(M)

[0063] Where V is wind speed, T is temperature, r RH is the relative humidity, M is the number of consecutive days without precipitation, f(V) is the quantitative value of the wind speed index, f(T) is the quantitative value of the temperature index, and f(r RH ) is the quantitative value of the relative humidity index, and f(M) is the quantitative value of the continuous precipitation-free day index.

[0064] In some embodiments, the device further includes a weight acquisition module, configured to:

[0065] The index weight of each index factor in the forest comprehensive fire risk assessment index system is obtained through the gray correlation method and the improved hierarchical analysis method.

[0066] In some embodiments, when the weight acquisition module obtains the index weight of each index factor in the forest fire risk assessment index system by using the gray correlation method and the improved analytic hierarchy process, it is used to:

[0067] Obtain historical forest burnt area data and corresponding historical data of each indicator factor in the forest comprehensive fire risk assessment indicator system;

[0068] The historical forest burnt area data is used as a reference sequence, and the corresponding historical data of each indicator factor in the forest comprehensive fire risk assessment index system is used as a comparison sequence;

[0069] Based on the comparison sequence and the reference sequence, calculating the correlation coefficient and correlation degree of each indicator factor in the comprehensive forest fire risk assessment index system;

[0070] Based on the correlation coefficient and correlation degree of each indicator factor in the forest fire risk comprehensive assessment indicator system, the correlation order of the indicator factors in the forest fire risk comprehensive assessment indicator system is obtained; the correlation order represents the correlation degree ranking;

[0071] Based on the index level of each index factor in the forest fire risk comprehensive assessment index system, a judgment matrix is ​​constructed; the judgment matrix is ​​used to compare the importance of two index factors in the same index level; the importance of the index factor is obtained based on the correlation sequence of the index factors in the forest fire risk comprehensive assessment index system and a preset scale scoring standard; the preset scale scoring standard is a scale table of 1 to 20;

[0072] Obtaining the maximum eigenvalue of the judgment matrix, and calculating a random consistency index based on the maximum eigenvalue;

[0073] When the random consistency index meets the consistency test threshold, the square root method is used to calculate the maximum eigenvalue and eigenvector of the judgment matrix; and the weight of each indicator factor in the comprehensive forest fire risk assessment index system is determined based on the eigenvector.

[0074] In some implementations, the level determination module is used to:

[0075] If P 综合火险 When the fire risk is less than 39.8, the corresponding comprehensive forest fire risk level is level one, which is a low fire risk.

[0076] If 39.8≤P 综合火险 When the fire risk is less than 47.2, the corresponding comprehensive forest fire risk level is Level 2, which is a relatively low fire risk.

[0077] If 47.2≤P 综合火险 When the value is less than 66.4, the corresponding comprehensive forest fire risk level is level 3, which is a relatively high fire risk;

[0078] If 66.4≤P 综合火险 When the temperature is less than 73, the corresponding comprehensive forest fire risk level is level 4, which is a high fire risk;

[0079] If 73≤P 综合火险 When it is less than 100, the corresponding comprehensive forest fire risk level is level 5, which is an extremely high fire risk.

[0080] In some embodiments, the fire risk assessment module is further used to:

[0081] In the case where fire point information is monitored in the forest area to be evaluated, determining a value of the fire point monitoring sensitivity;

[0082] The value of the fire point monitoring sensitivity, the value of the precipitation correction coefficient and the value of the snow accumulation correction coefficient are input into the comprehensive forest fire risk prediction model to obtain the comprehensive forest fire risk value of the forest area to be evaluated.

[0083] In some embodiments, the prediction data of the natural fire source in the future second time period includes lightning warning information; when the index acquisition module acquires the index quantization value of the natural fire source, it is used to:

[0084] Based on the lightning warning information and the lightning warning classification rules, a quantitative index value of the natural fire source is obtained;

[0085] When the indicator acquisition module performs quantization processing on the prediction data based on the quantization rule of the indicator factor to obtain the indicator quantization value of the indicator factor, it is used to:

[0086] Based on the classification rule of the indicator factor, determine the target category to which the predicted data of the indicator factor belongs;

[0087] Based on the category quantitative scoring rule of the indicator factor, the indicator quantitative value of the indicator factor is obtained.

[0088] To achieve the above-mentioned purpose, the third aspect embodiment of the present application proposes an electronic device, comprising: a processor, and a memory communicatively connected to the processor; the memory stores computer-executable instructions; the processor executes the computer-executable instructions stored in the memory to implement the method described in the first aspect.

[0089] To achieve the above-mentioned purpose, the fourth aspect of the present application proposes a computer-readable storage medium, in which computer-readable storage medium is stored computer-executable instructions, and when the computer-executable instructions are executed by a processor, they are used to implement the method described in the first aspect.

[0090] To achieve the above-mentioned purpose, the fifth aspect of the present application proposes a computer program product, including a computer program, which implements the method described in the first aspect when executed by a processor.

[0091] The comprehensive forest fire risk prediction method, device, electronic device and storage medium provided in the present application construct a comprehensive forest fire risk assessment index system including hazardous factors and exposure factors, integrates multiple index factors such as meteorological elements, forest combustible elements, terrain elements, fire source elements, exposure elements, etc., and establishes a comprehensive forest fire risk assessment model that covers the forest fire risk meteorological index and takes into account the sensitivity of fire point monitoring; obtains the indicator quantification value of each indicator factor of the comprehensive forest fire risk assessment index system and combines it with the comprehensive forest fire risk assessment model to realize the comprehensive forest fire risk prediction, promotes the refinement level of forest fire risk prediction, improves the comprehensiveness of forest fire risk prediction, and thus improves the accuracy of comprehensive forest fire risk prediction.

[0092] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through the practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0093] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0094] Figure 1 A schematic diagram of a flow chart of a forest fire comprehensive risk prediction method provided in an embodiment of the present application;

[0095] Figure 2 A schematic diagram of a comprehensive forest fire risk assessment index system provided in an embodiment of the present application;

[0096] Figure 3 A block diagram of a forest fire comprehensive risk prediction device provided in an embodiment of the present application;

[0097] Figure 4 A block diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0098] Embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present application, and should not be construed as limiting the present application.

[0099] The following describes the forest fire comprehensive risk prediction method, device and equipment of the embodiments of the present application with reference to the accompanying drawings.

[0100] Figure 1 A flow chart of a method for comprehensive forest fire risk prediction provided in an embodiment of the present application.

[0101] It should be noted that the executor of the forest fire comprehensive risk prediction method of the embodiment of the present application is the forest fire comprehensive risk prediction device of the embodiment of the present application, and the forest fire comprehensive risk prediction device can be configured in an electronic device so that the electronic device can perform the forest fire comprehensive risk prediction function.

[0102] like Figure 1 As shown, the forest fire comprehensive risk prediction method includes the following steps:

[0103] Step 101, based on the precipitation data and snow accumulation data of the forest area to be evaluated in the first time period, determine the value of the precipitation correction coefficient and the value of the snow accumulation correction coefficient.

[0104] It should be noted that the forest area to be assessed is the area where a comprehensive forest fire risk forecast is required.

[0105] In some embodiments, according to the correction coefficient of precipitation factor in GB / T 36743-2018 "Forest Fire Hazard Meteorological Level", "24h precipitation R r When ≥1mm, the precipitation correction coefficient C r =0; R r When <1mm, C r =1". Thus, the value of the precipitation correction coefficient is determined by the precipitation in the last 24 hours.

[0106] In some embodiments, according to the correction coefficient of snow accumulation factor in GB / T 36743-2018 "Forest Fire Danger Meteorological Level", "24h snow depth H s When >0cm, the snow accumulation correction factor C s =0; H s =0cm, C s =1". Thus, the value of the snow accumulation correction coefficient is determined based on the snow accumulation in the last 24 hours.

[0107] Step 102, determining whether fire point information is monitored in the forest area to be evaluated.

[0108] As an implementation method, it can be determined through drones or satellite images whether fire point information is monitored in the forest area to be evaluated.

[0109] Step 103, when no fire point information is monitored in the forest area to be evaluated, obtain the prediction data of each indicator factor in the forest comprehensive fire risk assessment index system of the forest area to be evaluated in the second time period in the future; and for the prediction data of each indicator factor, quantize the prediction data based on the quantization rule of the indicator factor to obtain the indicator quantization value of the indicator factor.

[0110] It should be noted that the embodiment of the present application takes into account the hazard index and exposure index to construct a comprehensive forest fire risk assessment index system.

[0111] In some embodiments, the first-level indicators of the comprehensive forest fire risk assessment index system include hazard indicators and exposure indicators, the second-level indicators of the hazard indicators include vegetation factors, meteorological factors, terrain factors and fire source factors, the tertiary indicators of the vegetation factors include vegetation type and EDVI double-layer vegetation index, the third-level indicators of the meteorological factors include precipitation, snow accumulation, wind speed, temperature, relative humidity and drought level, the third-level indicators of the terrain factors include altitude, slope and slope direction, the third-level indicators of the fire source factors include natural fire sources and man-made fire sources, and the fourth-level indicators of the man-made fire sources include the spatial distribution of settlements and road network density; the second-level indicators of the exposure indicators include population density and GDP spatial distribution.

[0112] like Figure 2 As shown in the figure, the comprehensive forest fire risk assessment index system is an integrated forest fire risk assessment index system that integrates multiple factors such as meteorology, vegetation, terrain, and fire sources. It includes two primary indicators: hazard index and exposure index.

[0113] The danger of forest fire is the possibility of disaster-causing effects, and the core is the activity level of the disaster-causing effects. The higher the activity level of the disaster-causing effects, the greater the danger and the more serious the losses from the disaster. The danger factors mainly consider the factors that affect the possibility of forest fires and the spread of fires, including the inherent risks of vegetation, fire source factors, the influence of climatic conditions and the influence of terrain. Therefore, the danger index in the present invention includes four secondary indicators: vegetation factors, meteorological factors, terrain factors and fire source factors.

[0114] Forest fire exposure is the exposure vulnerability of various disaster-bearing carriers that may be threatened by dangerous factors in the evaluation area. The exposure risk factor considers the impact on the disaster-bearing body, mainly evaluating all people and properties that may be threatened by fire risks in the evaluation area. The more people and properties exposed to the disaster threat, the greater the disaster risk, including factors such as population density and GDP spatial distribution in the evaluation area. Therefore, the exposure index in the present invention includes two secondary indicators: population density and GDP spatial distribution.

[0115] Forest vegetation is complex and diverse, and the structure of the forest stand and the characteristics of the vegetation community affect the occurrence and development of fires. Different vegetation types have different burning hazards. Generally, the fire resistance of coniferous forests, mixed coniferous and broad-leaved forests, deciduous broad-leaved forests, and evergreen broad-leaved forests gradually increases, and the flammability decreases successively. In addition to vegetation types, different vegetation moisture contents also have different fire hazards. The vegetation index (EDVI) is a characteristic parameter that displays the moisture content of vegetation based on the difference in the microwave emissivity of the two channels of 19GHz and 37GHz, and compared with the commonly used NDVI normalized vegetation index, it increases the consideration of the influence of vertical stratification of vegetation. A high value of EDVI corresponds to a high moisture content of vegetation, so a fire is not easy to occur, and even if a fire occurs, its burning intensity is relatively low. Therefore, the vegetation factors in the present invention include two three-level indicators: vegetation type and EDVI double-layer vegetation index.

[0116] The occurrence of fire is closely related to climate change. The increase in temperature, extension of drought period and decrease in air humidity caused by climate change will lead to the advancement and extension of the fire risk period, the increase in the burned area and the increase in fire intensity. The present invention mainly selects four factors as climate condition factors: temperature, precipitation (snow), wind force and relative air humidity. With reference to the national standard GB / T36743-2018 "Forest Fire Risk Meteorological Level", the meteorological factors in the present invention include six three-level indicators: precipitation, snow accumulation, wind speed, temperature, relative humidity and drought level.

[0117] Topography is an important fire environment factor, and its impact on forest fires is very obvious. The undulating topography forms different fire environments, which not only affect the occurrence and development of fires, but also directly affect the spread and intensity of fires. Altitude has a significant impact on the zonal distribution of plants. With the change of altitude, the meteorological and climatic characteristics also change accordingly. The higher the altitude, the temperature will gradually drop, and the possibility of fire will decrease. The slope directly affects the amount of solar radiation received by the surface. The more heat the sunny slope receives, the lower the moisture content of the vegetation, and the greater the possibility of fire. The slope determines the residence time of water on the ground. The steeper the slope, the easier it is for water to lose, and the combustibles are very easy to dry, and the greater the danger. Here, the three factors of altitude, slope, and slope are mainly selected as terrain influencing factors. Therefore, the terrain factors in the present invention include three three-level indicators: altitude, slope, and slope.

[0118] Fire sources include man-made fire sources and natural fire sources. Natural fire sources mainly include lightning strikes and the like. Man-made fire sources include forest fires caused by careless use of fire in daily life, fires caused by fire in agricultural, forestry and animal husbandry production, fires used for burning paper at graves during the Qingming Festival, and fires used for setting off firecrackers. The existence of fire sources is closely related to human behavior, and residential areas and their activity places are high-incidence areas for fire sources. Human activities are one of the main causes of forest fires. The existence of man-made fire sources is affected by multiple factors such as the distance from residential areas and the density of road traffic networks. Here, the spatial distribution of the distance from residential areas and the spatial distribution of the distance from roads are mainly selected as fire source influencing factors. Therefore, in the present invention, natural fire sources mainly consider lightning strike warnings, and man-made fire sources include two four-level indicators: spatial distribution of residential areas and road network density.

[0119] As an implementation method, a method for obtaining the predicted data of each indicator factor in the forest comprehensive fire risk assessment index system of the forest area to be assessed in the second time period in the future includes: obtaining the predicted data of each indicator in the forest comprehensive fire risk assessment index system for the next week or longer through different methods such as network crawling or direct data access.

[0120] In some embodiments, the prediction data of the natural fire source in the second time period in the future includes lightning warning information; the method for obtaining the index quantification value of the natural fire source includes: obtaining the index quantification value of the natural fire source based on the lightning warning information and the lightning warning classification rules.

[0121] As an implementation method, the implementation method of obtaining the indicator quantization value of each indicator factor includes:

[0122] For each indicator factor, based on the category classification rule of the indicator factor, determine the target category to which the predicted data of the indicator factor belongs; based on the category quantitative scoring rule of the indicator factor, obtain the indicator quantitative value of the indicator factor.

[0123] For example, for 1) vegetation type:

[0124] Vegetation types formed under different terrain and climatic conditions have different characteristics, resulting in differences in the form, quantity, extracts and ash content of the combustibles formed, which in turn affects the ignition point and calorific value of the combustibles. Different vegetation types have different flammability, which also determines different dangers, that is, the greater the flammability of the vegetation, the higher the danger.

[0125] With reference to the 1:1000000 China Vegetation Atlas edited by Academician Hou Xueyu, a famous vegetation ecologist, it reflects in detail the distribution conditions, horizontal zonality and vertical zonality distribution laws of 796 vegetation units of 11 vegetation type groups and 54 vegetation types in my country, and the vegetation types in the study area are divided into coniferous forests, mixed coniferous and broad-leaved forests, deciduous broad-leaved forests, broad-leaved forests, etc., based on the major vegetation categories as the classification standards. In general, the fire resistance of coniferous forests, mixed coniferous and broad-leaved forests, broad-leaved forests, etc. gradually increases, and the flammability gradually decreases. According to the classification of vegetation types and the ranking of danger, the present invention has quantitatively scored different vegetation classes, and the scores are as shown in Table 1.

[0126] Table 1:

[0127]

[0128]

[0129] For example, if the vegetation type of the forest area to be assessed is mainly coniferous forest, the quantitative value of the vegetation type indicator is 80.

[0130] 2) Vegetation Index

[0131] The EDVI double-layer vegetation index is also one of the important factors affecting the occurrence and spread of fire. 10-18 and EDVI 18-37 ) characterizes the moisture content and fuel load of the vegetation underlayer and vegetation canopy. According to relevant knowledge, in the time series changes, EDVI 18-37 It has a negative correlation with the number of fire points FC and the fire radiation power FRP on a time scale; EDVI 10-18 There is also a negative correlation between it and the number of fire points NOF and fire radiation energy FRE.

[0132] Based on the above research, the EDVI double-layer vegetation index was quantified. In areas with lush vegetation, the vegetation index ranged from (-0.02 to 0.02), mainly concentrated between (0.008 and 0.01). As the EDVI value increased, the number of fire points decreased. The quantitative scores of indicators of different categories of vegetation index are shown in Table 2.

[0133] Table 2:

[0134]

[0135] 3) Wind speed

[0136] According to the classification of wind speed in GB / T 36743-2018 "Forest Fire Hazard Meteorological Level", the quantitative scoring of the indicators is shown in Table 3.

[0137] Table 3:

[0138]

[0139]

[0140] 4) Temperature

[0141] According to the temperature classification in GB / T 36743-2018 "Forest Fire Hazard Meteorological Level", the quantitative scoring of the indicators is shown in Table 4.

[0142] Table 4:

[0143]

[0144] 5) Relative humidity

[0145] According to the classification of relative humidity in GB / T 36743-2018 "Forest Fire Hazard Meteorological Level", the quantitative scoring of the indicators is shown in Table 5.

[0146] Table 5:

[0147]

[0148] 6) Drought level

[0149] According to the classification of drought levels in GB / T 36743-2018 "Forest Fire Hazard Meteorological Levels", the quantitative scoring table of indicators is 6.

[0150] Table 6:

[0151]

[0152]

[0153] 7) Altitude

[0154] Altitude has a profound impact on forest fire risk. The present invention divides the altitude into six levels: >900m, 800-900m, 700-800m, 600-700m, 500-600m, and 0-500m. The quantitative scoring of the indicators is shown in Table 7.

[0155] Table 7:

[0156]

[0157] 8) Slope

[0158] Slope is one of the most important topographic factors in forest fire risk assessment, and has a significant impact on the occurrence and spread of fire. The present invention divides the slope into five levels: >35°, 25-35°, 15-25°, 5-15°, and 0-5°. The quantitative scoring of the indicators is shown in Table 8.

[0159] Table 8:

[0160]

[0161] 9) Slope direction

[0162] The slope direction directly affects the amount of solar radiation received by the surface, and there is a certain relationship between it and the forest fire risk. Different slope directions may lead to different fire risks. The present invention divides the slope direction into four levels, and the danger ranking is sunny slope > semi-sunny slope > semi-shady slope > shady slope. The quantitative scoring of the indicators is shown in Table 9.

[0163] Table 9:

[0164]

[0165] 10) Natural fire sources

[0166] Natural fire sources mainly consider lightning warning. According to the meteorological classification of lightning warning, lightning index factors are divided into three levels according to the degree of harm, namely,

[0167] Level 1 Yellow Warning: Lightning activity may occur within 6 hours, which may cause lightning disasters;

[0168] Level 2 Orange Warning: There is a high possibility of lightning activity within 2 hours, or the area has been affected by lightning activity and may continue, and there is a high possibility of lightning disaster accidents;

[0169] Level 3 Red Alert: There is a high possibility of lightning activity within 2 hours, or strong lightning activity has already occurred and may continue, and the possibility of a lightning disaster accident is very high.

[0170] According to the lightning warning classification, the quantitative scoring of the natural fire source index of the present invention is shown in Table 10.

[0171] Table 10:

[0172]

[0173] 11) Spatial distribution of settlements

[0174] There is a close correlation between the spatial distribution of settlements and the risk of forest fires. Generally speaking, the closer a settlement is to a forest fire, the more human activities will increase the possibility of causing a fire, and the higher the risk of forest fires faced by the area where it is located. The more settlements there are in a unit space, the greater the danger. The quantitative scoring of the indicators in this invention is shown in Table 11.

[0175] Table 11:

[0176]

[0177]

[0178] 12) Road network density

[0179] Where the road network density is high, human activities are concentrated, vehicle traffic is large, potential fire sources increase, and the surrounding vegetation is more likely to burn. Referring to LY / T 1063-2008 "National Forest Fire Hazard Zoning Grade", the road network density is divided into three levels. The quantitative scoring of the indicators of the present invention is shown in Table 12.

[0180] Table 12:

[0181]

[0182] 13) Population density

[0183] LY / T 1063-2008 National Forest Fire Hazard Zoning Classification divides population density into three levels. The quantitative scoring of the indicators in the present invention is shown in Table 13.

[0184] Table 13:

[0185]

[0186] 14) Spatial distribution of GDP

[0187] According to the GDP per unit area, the higher the GDP, the more developed the economy, the larger the population, and the greater the fire hazard. The quantitative scoring of the indicators in the present invention is shown in Table 14.

[0188] Table 14:

[0189]

[0190] Therefore, the quantitative value of each indicator factor can be obtained based on the predicted data of each indicator factor.

[0191] Step 104, inputting the index quantification value of each index factor in the comprehensive forest fire risk assessment index system, the value of the precipitation correction coefficient and the value of the snow accumulation correction coefficient into the comprehensive forest fire risk prediction model to obtain the comprehensive forest fire risk value of the forest area to be assessed.

[0192] In some embodiments, the forest fire risk prediction model is expressed as follows:

[0193]

[0194] Among them, P 综合火险 is the comprehensive forest fire risk value, C r is the precipitation correction coefficient. When the 24h precipitation is greater than or equal to 1mm, the value is 0; when the 24h precipitation is less than 1mm, the value is 1; Cs is the snow depth correction coefficient. When the 24-hour snow depth is greater than 0, the value is 0; when the 24-hour snow depth is equal to 0, the value is 1; U FFDI W is the forest fire weather index; FFDI is the index weight of meteorological factors; A i is the quantitative value of the ith indicator, excluding meteorological factors, including 10 indicators: vegetation type, vegetation index, altitude, slope, slope aspect, natural fire source, spatial distribution of distance to residential areas, road network density, population density, and GDP spatial distribution; W i is the indicator weight of the ith indicator, excluding meteorological factors; n is the number of evaluation indicators, excluding meteorological factors; K is the fire point monitoring sensitivity. When a fire point is detected, the area where the fire point is located is directly considered to be extremely high risk, with a value of 100; if no fire point is detected, the value is 0.

[0195] Among them, the expression of forest fire weather index is as follows:

[0196] U FFDI =f(V)+f(T)+f(r RH )+f(M)

[0197] Where V is wind speed, T is temperature, r RH is the relative humidity, M is the number of consecutive days without precipitation, f(V) is the quantitative value of the wind speed index, f(T) is the quantitative value of the temperature index, and f(r RH ) is the quantitative value of the relative humidity index, and f(M) is the quantitative value of the continuous precipitation-free day index.

[0198] In some embodiments, a method for obtaining the indicator weight of each indicator factor in the comprehensive forest fire risk assessment indicator system includes:

[0199] The index weight of each index factor in the forest comprehensive fire risk assessment index system is obtained through the gray correlation method and the improved hierarchical analysis method.

[0200] In some embodiments, the index weights of each index factor in the forest fire risk assessment index system are obtained by using the gray correlation method and the improved analytic hierarchy process, including:

[0201] Obtain historical forest burnt area data and corresponding historical data of each indicator factor in the forest comprehensive fire risk assessment indicator system;

[0202] The historical forest burnt area data is used as a reference sequence, and the corresponding historical data of each indicator factor in the forest comprehensive fire risk assessment index system is used as a comparison sequence;

[0203] Based on the comparison sequence and the reference sequence, calculating the correlation coefficient and correlation degree of each indicator factor in the comprehensive forest fire risk assessment index system;

[0204] Based on the correlation coefficient and correlation degree of each indicator factor in the forest fire risk assessment index system, the correlation order of the indicator factors in the forest fire risk assessment index system is obtained; the correlation order represents the correlation degree ranking;

[0205] Based on the index level of each index factor in the forest fire risk comprehensive assessment index system, a judgment matrix is ​​constructed; the judgment matrix is ​​used to compare the importance of two index factors in the same index level; the importance of the index factor is obtained based on the correlation sequence of the index factors in the forest fire risk comprehensive assessment index system and a preset scale scoring standard; the preset scale scoring standard is a scale table of 1 to 20;

[0206] Obtaining the maximum eigenvalue of the judgment matrix, and calculating a random consistency index based on the maximum eigenvalue;

[0207] When the random consistency index meets the consistency test threshold, the square root method is used to calculate the maximum eigenvalue and eigenvector of the judgment matrix; and the weight of each indicator factor in the comprehensive forest fire risk assessment index system is determined based on the eigenvector.

[0208] It can be understood that the correlation coefficient and correlation degree of each risk indicator factor are calculated through historical data (including historical fire point data and historical data of each indicator in the corresponding evaluation index system), and the impact ranking of each indicator factor on the degree of forest fire is analyzed; then based on the hierarchical analysis method, a hierarchical matrix is ​​constructed, and the 1-40 scaling method proposed in the present invention is used to perform expert scoring and calculate the impact weight of each indicator. Further, the grayscale correlation method is used to construct a comparison matrix through the historical burned area data and the corresponding historical data of each indicator of the index system constructed by this method to objectively analyze the importance of each indicator factor on forest fire risk; then the hierarchical analysis method is used to perform indicator factor weight analysis, wherein the commonly used scale of the hierarchical analysis method is to use the 1-9 scaling method to construct the comparison matrix. In this method, due to the large number of indicator factors and the wider range of indicator categories covered, in order to avoid the consistency problem of the comparison matrix constructed when comparing two indicator factors, the scale value is expanded, and the 1-20 scaling method in Table 15 is used to construct the comparison matrix.

[0209] Table 15:

[0210]

[0211] Exemplary:

[0212] Based on the historical burned area data and the corresponding historical data of each indicator of the indicator system constructed by this method, the process of calculating the weight influence of each indicator factor is as follows:

[0213] 1) Determine the analysis sequence

[0214] Determine the reference sequence that reflects the system behavior characteristics and the comparison sequence that affects the system behavior. The data sequence that reflects the system behavior characteristics is called the reference sequence. The data sequence composed of factors that affect the system behavior is called the comparison sequence.

[0215] In this method, the historical data of indicators corresponding to each indicator factor in the comprehensive forest fire risk indicator system are taken as a column to form a comparison sequence; the forest burned area data are taken as a reference sequence.

[0216] The reference sequence (also called the mother sequence) is Y = {Y(k) | k = 1, 2, 3...n};

[0217] The comparison sequence (also called subsequence) is X i ={X i (k)|k=1,2,3...n}(i=1,2,3...m).

[0218] This example selects forest fire area data from 6 places and obtains data on various risk indicators in the corresponding forest fire risk index system, forming the analysis series shown in Table 16 below:

[0219] Table 16:

[0220]

[0221] 2) Dimensionless standardization of indicator series

[0222] Since the data in each factor column of the system may be inconvenient to compare or it is difficult to get the correct conclusion when comparing due to different dimensions. Therefore, when performing grey correlation analysis, dimensionless standardization of the data is required. The qualitative indicators are quantified using the divided grade values, and then the series is transformed to make the series dimensionless data to achieve comparability. Common methods include: initialization, averaging, min-max standardization. The present invention adopts dimensionless standardization of averaging each indicator.

[0223] The historical data obtained above are dimensionlessly normalized to form a standardized sequence as shown in Table 17.

[0224] Table 17:

[0225]

[0226] 3) Calculate the correlation coefficient and correlation degree

[0227] The correlation coefficient between the comparison series corresponding to each indicator factor and the reference series is calculated as follows:

[0228]

[0229] Δ i (k)=|Y(k)-X i (k)|

[0230] Among them, ξ i (k) represents the correlation coefficient between the kth parameter of the i-th subsequence and the kth parameter of the parent sequence. ρ is called the resolution coefficient. The smaller ρ is, the greater the resolution is. Generally, the value range of ρ is (0, 1). The specific value can be determined according to the situation. When ρ≤0.5463, the resolution is the best. This method takes ρ=0.5.

[0231] The correlation coefficient indicates the degree of correlation between each data column. The average value is used to calculate the correlation degree, which is shown as follows:

[0232]

[0233] Correlation is a measure of the correlation between the index factors. When comparing the correlation between two indexes, if X i The correlation is greater than X j If the correlation degree is , it means that the influence of the i-th indicator factor on forest fire risk is greater than that of the j-th indicator factor. Sorting by the size of the correlation degree will give the correlation order. The size of the correlation order reflects the influence of the comparison factor on the reference factor.

[0234] Based on historical data, the present invention calculates the correlation between each indicator factor, as shown in Table 18 below.

[0235] Table 18:

[0236]

[0237] The correlation degree and correlation order of each indicator factor obtained are shown in Table 19.

[0238] Table 19:

[0239]

[0240]

[0241] It can be seen that, combined with historical data, the correlation order of each indicator factor from high to low is: temperature, drought level, wind speed, vegetation type, relative humidity, spatial distribution of distance from residential areas, precipitation, population density, spatial distribution of GDP, snow accumulation, slope, natural fire source (lightning warning), EDVI double-layer vegetation index, road network density, slope aspect, and altitude.

[0242] (4) Combined with the obtained index factor correlation order (the degree of importance of forest fire), the weight of each index factor is calculated as follows:

[0243] 1) Establish a hierarchical structure of the comprehensive forest fire risk assessment indicator system, see Table 20.

[0244] Table 20:

[0245]

[0246]

[0247] 2) Based on the obtained indicator factor correlation sequence, the 1-40 scaling method in Table 15 and the expert scoring, a judgment matrix is ​​constructed, that is, the indicator factors at the same level are compared pairwise to form a judgment matrix.

[0248] A=(a ij ) n×n

[0249] In the formula, A represents the n-order judgment matrix, n represents the number of indicators, and a i j represents the importance of the i-th indicator relative to the j-th indicator.

[0250] Different from the commonly used hierarchical analysis method, there is no need to rank the importance of indicators here. You can directly refer to the indicator factor correlation sequence results obtained by the coupling weight analysis method and have experts score them. You only need to give a judgment matrix based on the 1-40 scale scoring standard.

[0251] 3) Matrix consistency check: Calculate the maximum characteristic root λ of the judgment matrix A max , using λ max Calculate the random consistency index CR. If CR < 0.1, the matrix meets the consistency requirements.

[0252] in

[0253] Where CR is the consistency ratio, CI is the consistency index, RI is the corresponding random consistency index, and n is the order of the judgment matrix.

[0254] 4) Single-level weight sorting.

[0255] The maximum eigenvalue λ is calculated using the square root method max and the eigenvector w i , first multiply the elements of A row by row, and then take the nth root of the products, that is,

[0256]

[0257] The square root vector After normalization, w i That is the desired eigenvector, which is the weight of the index of this layer:

[0258]

[0259] 5) Comprehensive evaluation ranking weight. The comprehensive weight refers to the comprehensive weight of the factor layer index to the target layer. The principle is that the factor layer index weight is transferred through the corresponding parent node index weight of each layer in the criterion layer to obtain the target layer weight. If the comprehensive weight of the k-th layer factor index to the target layer is known to be in The weight of the i-th factor in the k+1th layer to the j-th factor in the kth layer is Then the comprehensive weight of the k+1th layer factor to the target is:

[0260] Where i=1,2,3…n k+1

[0261] The weights of each indicator factor are shown in Table 21.

[0262] Table 21:

[0263]

[0264]

[0265] This embodiment is based on the coupled weight analysis method of grayscale correlation and hierarchical analysis method. It relies on objective historical data to obtain the correlation order of each indicator factor based on the grayscale correlation method, that is, the order of the impact of the indicator factors on the fire risk, and then uses the improved scaling method based on the hierarchical analysis method to perform expert scoring to calculate the weight. This coupled weight analysis method has the characteristics of quantitative analysis of the grayscale correlation method, and also has the advantages of a wide range of applications of the hierarchical analysis method. It avoids the subjective problem of weight analysis caused by the hierarchical analysis method relying on expert scoring, and also avoids the problem of inaccurate weight calculation due to insufficient data sample size in the grayscale correlation method.

[0266] Therefore, the quantified index values ​​of each index factor in the comprehensive forest fire risk assessment index system, the value of the precipitation correction coefficient and the value of the snow accumulation correction coefficient are input into the comprehensive forest fire risk prediction model to obtain the comprehensive forest fire risk value of the forest area to be assessed.

[0267] The above scheme is implemented for the case where no fire point is detected. When the fire point information is detected, the comprehensive forest fire risk value of the forest area to be evaluated is obtained by the following method.

[0268] In some embodiments, when fire point information is monitored in the forest area to be evaluated, the value of the fire point monitoring sensitivity is determined; the value of the fire point monitoring sensitivity, the value of the precipitation correction coefficient and the value of the snow accumulation correction coefficient are input into the comprehensive forest fire risk prediction model to obtain the comprehensive forest fire risk value of the forest area to be evaluated.

[0269] Therefore, when a fire point is monitored in the forest area to be evaluated, the value of the direct fire point monitoring sensitivity, the value of the precipitation correction coefficient and the value of the snow accumulation correction coefficient are input into the comprehensive forest fire risk prediction model to obtain the comprehensive forest fire risk value of the forest area to be evaluated; there is no need to obtain the predicted data of each indicator factor in the comprehensive forest fire risk assessment index system.

[0270] Step 105, based on the comprehensive forest fire risk value of the forest area to be evaluated and the forest fire risk level classification threshold, obtain the comprehensive forest fire risk level of the forest area to be evaluated.

[0271] As an implementation method, obtaining the comprehensive forest fire risk level of the forest area to be assessed includes:

[0272] If P 综合火险 When the fire risk is less than 39.8, the corresponding comprehensive forest fire risk level is level one, which is a low fire risk.

[0273] If 39.8≤P 综合火险 When the fire risk is less than 47.2, the corresponding comprehensive forest fire risk level is Level 2, which is a relatively low fire risk.

[0274] If 47.2≤P 综合火险 When the value is less than 66.4, the corresponding comprehensive forest fire risk level is level 3, which is a relatively high fire risk;

[0275] If 66.4≤P 综合火险 When the temperature is less than 73, the corresponding comprehensive forest fire risk level is level 4, which is a high fire risk;

[0276] If 73≤P 综合火险 When it is less than 100, the corresponding comprehensive forest fire risk level is level 5, which is an extremely high fire risk.

[0277] Therefore, the comprehensive forest fire risk level is determined according to the forest fire risk classification standards.

[0278] The forest fire comprehensive risk prediction method of the embodiment of the present application constructs a forest fire comprehensive risk assessment index system including hazardous factors and exposure factors, integrates multiple index factors such as meteorological factors, forest combustible factors, terrain factors, fire source factors, and exposure factors, and establishes a forest fire comprehensive risk assessment model that covers forest fire meteorological indexes and considers the sensitivity of fire point monitoring; obtains the index quantification value of each index factor of the forest fire comprehensive risk assessment index system and combines it with the forest fire comprehensive risk assessment model to achieve forest fire comprehensive risk prediction, promotes the refinement level of forest fire risk prediction, and improves the comprehensiveness of forest fire risk prediction, thereby improving the accuracy of forest fire comprehensive risk prediction. In addition, a coupling weight analysis method based on gray correlation and hierarchical analysis method relies on objective historical data to obtain the correlation order of each index factor based on the gray correlation method, and uses the improved scaling method based on the hierarchical analysis method to perform expert scoring, thereby calculating the influence weight of each index factor. The coupled weight analysis method has the characteristics of quantitative analysis of the grayscale correlation method, and also has the advantage of a wide range of applications of the hierarchical analysis method. It avoids the subjective problem of weight analysis caused by the hierarchical analysis method relying on expert scoring, and also avoids the problem of inaccurate weight calculation due to insufficient data sample size in the grayscale correlation method.

[0279] In order to implement the above-mentioned embodiments, the present application also proposes a comprehensive forest fire risk prediction device. Figure 3 This is a schematic diagram of the structure of a forest fire comprehensive risk prediction device provided in an embodiment of the present application. Figure 3 As shown, the forest fire comprehensive risk prediction device may include: a coefficient determination module 301, a fire point acquisition module 302, an index acquisition module 303, a fire risk assessment module 304 and a level determination module 305.

[0280] The coefficient determination module 301 is used to determine the value of the precipitation correction coefficient and the value of the snow accumulation correction coefficient based on the precipitation data and the snow accumulation data of the forest area to be evaluated in the first time period;

[0281] A fire point acquisition module 302 is used to determine whether fire point information is monitored in the forest area to be evaluated;

[0282] The indicator acquisition module 303 is used to obtain the forecast data of each indicator factor in the forest comprehensive fire risk assessment indicator system of the forest area to be assessed in the future second time period when no fire point information is monitored in the forest area to be assessed; and for the forecast data of each indicator factor, quantize the forecast data based on the quantization rule of the indicator factor to obtain the indicator quantization value of the indicator factor;

[0283] The fire risk assessment module 304 is used to input the index quantification value of each index factor in the forest comprehensive fire risk assessment index system, the value of the precipitation correction coefficient and the value of the snow accumulation correction coefficient into the forest comprehensive fire risk prediction model to obtain the forest comprehensive fire risk value of the forest area to be assessed;

[0284] The level determination module 305 is used to obtain the comprehensive forest fire risk level of the forest area to be evaluated based on the comprehensive forest fire risk value of the forest area to be evaluated and the forest fire risk level classification threshold.

[0285] Furthermore, in a possible implementation of the embodiment of the present application, the primary index of the comprehensive forest fire risk assessment index system includes a hazard index and an exposure index.

[0286] The secondary indicators of the hazard index include vegetation factors, meteorological factors, terrain factors and fire source factors. The tertiary indicators of the vegetation factors include vegetation type and EDVI double-layer vegetation index. The tertiary indicators of the meteorological factors include precipitation, snow accumulation, wind speed, temperature, relative humidity and drought level. The tertiary indicators of the terrain factors include altitude, slope and slope direction. The tertiary indicators of the fire source factors include natural fire sources and man-made fire sources. The quaternary indicators of man-made fire sources include the spatial distribution of settlements and road network density.

[0287] The secondary indicators of the exposure index include population density and GDP spatial distribution.

[0288] Furthermore, in a possible implementation of the embodiment of the present application, the forest comprehensive fire risk prediction model is expressed as follows:

[0289]

[0290] Among them, P 综合火险 is the comprehensive forest fire risk value, C r is the precipitation correction coefficient. When the 24h precipitation is greater than or equal to 1mm, the value is 0; when the 24h precipitation is less than 1mm, the value is 1; C s is the snow depth correction coefficient. When the 24-hour snow depth is greater than 0, the value is 0; when the 24-hour snow depth is equal to 0, the value is 1; U FFDI W is the forest fire weather index; FFDI is the index weight of meteorological factors; A i is the quantitative value of the ith indicator, excluding meteorological factors, including 10 indicators: vegetation type, vegetation index, altitude, slope, slope aspect, natural fire source, spatial distribution of distance to residential areas, road network density, population density, and GDP spatial distribution; W iis the indicator weight of the ith indicator, excluding meteorological factors; n is the number of evaluation indicators, excluding meteorological factors; K is the fire point monitoring sensitivity. When a fire point is detected, the area where the fire point is located is directly considered to be extremely high risk, with a value of 100; if no fire point is detected, the value is 0;

[0291] The forest fire risk meteorological index is expressed as follows:

[0292] U FFDI =f(V)+f(T)+f(r RH )+f(M)

[0293] Where V is wind speed, T is temperature, r RH is the relative humidity, M is the number of consecutive days without precipitation, f(V) is the quantitative value of the wind speed index, f(T) is the quantitative value of the temperature index, and f(r RH ) is the quantitative value of the relative humidity index, and f(M) is the quantitative value of the continuous precipitation-free day index.

[0294] Furthermore, in a possible implementation of the embodiment of the present application, the apparatus further includes a weight acquisition module 306, which is used to:

[0295] The index weight of each index factor in the forest comprehensive fire risk assessment index system is obtained through the gray correlation method and the improved hierarchical analysis method.

[0296] Furthermore, in a possible implementation of the embodiment of the present application, when the weight acquisition module 306 obtains the index weight of each index factor in the forest fire risk assessment index system by using the gray correlation method and the improved analytic hierarchy process, it is used to:

[0297] Obtain historical forest burnt area data and corresponding historical data of each indicator factor in the forest comprehensive fire risk assessment indicator system;

[0298] The historical forest burnt area data is used as a reference sequence, and the corresponding historical data of each indicator factor in the forest comprehensive fire risk assessment index system is used as a comparison sequence;

[0299] Based on the comparison sequence and the reference sequence, calculating the correlation coefficient and correlation degree of each indicator factor in the comprehensive forest fire risk assessment index system;

[0300] Based on the correlation coefficient and correlation degree of each indicator factor in the forest fire risk comprehensive assessment indicator system, the correlation order of the indicator factors in the forest fire risk comprehensive assessment indicator system is obtained; the correlation order represents the correlation degree ranking;

[0301] Based on the index level of each index factor in the forest fire risk comprehensive assessment index system, a judgment matrix is ​​constructed; the judgment matrix is ​​used to compare the importance of two index factors in the same index level; the importance of the index factor is obtained based on the correlation sequence of the index factors in the forest fire risk comprehensive assessment index system and a preset scale scoring standard; the preset scale scoring standard is a scale table of 1 to 20;

[0302] Obtaining the maximum eigenvalue of the judgment matrix, and calculating a random consistency index based on the maximum eigenvalue;

[0303] When the random consistency index meets the consistency test threshold, the square root method is used to calculate the maximum eigenvalue and eigenvector of the judgment matrix; and the weight of each indicator factor in the comprehensive forest fire risk assessment index system is determined based on the eigenvector.

[0304] Furthermore, in a possible implementation of the embodiment of the present application, the level determination module 305 is used to:

[0305] If P 综合火险 When the fire risk is less than 39.8, the corresponding comprehensive forest fire risk level is level one, which is a low fire risk.

[0306] If 39.8≤P 综合火险 When the fire risk is less than 47.2, the corresponding comprehensive forest fire risk level is Level 2, which is a relatively low fire risk.

[0307] If 47.2≤P 综合火险 When the value is less than 66.4, the corresponding comprehensive forest fire risk level is level 3, which is a relatively high fire risk;

[0308] If 66.4≤P 综合火险 When the temperature is less than 73, the corresponding comprehensive forest fire risk level is level 4, which is a high fire risk;

[0309] If 73≤P 综合火险 When it is less than 100, the corresponding comprehensive forest fire risk level is level 5, which is an extremely high fire risk.

[0310] Furthermore, in a possible implementation of the embodiment of the present application, the fire risk assessment module 304 is further configured to:

[0311] In the case where fire point information is monitored in the forest area to be evaluated, determining a value of the fire point monitoring sensitivity;

[0312] The value of the fire point monitoring sensitivity, the value of the precipitation correction coefficient and the value of the snow accumulation correction coefficient are input into the comprehensive forest fire risk prediction model to obtain the comprehensive forest fire risk value of the forest area to be evaluated.

[0313] Further, in a possible implementation of the embodiment of the present application, the prediction data of the natural fire source in the second future time period includes lightning warning information; when the index acquisition module 303 acquires the index quantization value of the natural fire source, it is used to:

[0314] Based on the lightning warning information and the lightning warning classification rules, a quantitative index value of the natural fire source is obtained;

[0315] When the indicator acquisition module 303 performs quantization processing on the prediction data based on the quantization rule of the indicator factor to obtain the indicator quantization value of the indicator factor, it is used to:

[0316] Based on the classification rule of the indicator factor, determine the target category to which the predicted data of the indicator factor belongs;

[0317] Based on the category quantitative scoring rule of the indicator factor, the indicator quantitative value of the indicator factor is obtained.

[0318] It should be noted that the aforementioned explanation of the forest fire comprehensive risk prediction method embodiment is also applicable to the forest fire comprehensive risk prediction device of this embodiment, and will not be repeated here.

[0319] In order to implement the above embodiment, the present application also proposes an electronic device. Figure 4 , Figure 4 Schematic diagram of the structure of the electronic device provided in the embodiment of the present application. Figure 4 As shown, the electronic device 400 includes: a processor 401, and a memory 402 communicatively connected to the processor 401; the memory 402 stores computer-executable instructions; the processor 401 executes the computer-executable instructions stored in the memory to implement the method provided in the aforementioned embodiment.

[0320] In order to implement the above embodiments, the present application also proposes a computer-readable storage medium, in which computer-executable instructions are stored. When the computer-executable instructions are executed by a processor, they are used to implement the methods provided by the above embodiments.

[0321] In order to implement the above embodiments, the present application also proposes a computer program product, including a computer program, which implements the methods provided by the above embodiments when executed by a processor.

[0322] The collection, storage, use, processing, transmission, provision and disclosure of user personal information involved in this application are in compliance with relevant laws and regulations and do not violate public order and good morals.

[0323] It should be noted that personal information from users should be collected for legitimate and reasonable purposes and should not be shared or sold outside of these legitimate uses. In addition, such collection / sharing should be carried out after receiving the user's informed consent, including but not limited to notifying the user to read the user agreement / user notice and sign the agreement / authorization including authorization of relevant user information before the user uses the function. In addition, any necessary steps should be taken to protect and safeguard access to such personal information data and ensure that others who have access to personal information data comply with its privacy policy and procedures.

[0324] The present application is expected to provide an implementation scheme for users to selectively block the use or access of personal information data. That is, the present disclosure is expected to provide hardware and / or software to prevent or block access to such personal information data. Once the personal information data is no longer needed, the risk can be minimized by limiting data collection and deleting the data. In addition, when applicable, such personal information is de-identified to protect the privacy of the user.

[0325] In the description of the aforementioned embodiments, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0326] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In the description of this application, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0327] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, fragment or portion of code comprising one or more executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present application belong.

[0328] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute the instructions), or in combination with these instruction execution systems, devices or apparatuses. For the purpose of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in combination with these instruction execution systems, devices or apparatuses. More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting or processing in other suitable ways if necessary, and then stored in a computer memory.

[0329] It should be understood that the various parts of the present application can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0330] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.

[0331] In addition, each functional unit in each embodiment of the present application may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0332] The storage medium mentioned above may be a read-only memory, a magnetic disk or an optical disk, etc. Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limiting the present application. A person of ordinary skill in the art may change, modify, replace and modify the above embodiments within the scope of the present application.

Claims

1. A forest fire comprehensive risk prediction method, characterized in that: The following steps are involved: Determine the value of the precipitation correction factor and the value of the snow accumulation correction factor based on the precipitation data and the snow accumulation data of the forest area to be evaluated in the first time period; Determining whether fire point information is monitored in the forest area to be assessed; In the case that no fire point information is monitored in the forest area to be evaluated, obtaining forecast data of each indicator factor in the forest comprehensive fire risk assessment index system of the forest area to be evaluated in the second time period in the future; And for the predicted data of each indicator factor, the predicted data is quantized based on the quantization rule of the indicator factor to obtain the indicator quantization value of the indicator factor; Inputting the index quantification value of each index factor in the comprehensive forest fire risk assessment index system, the value of the precipitation correction coefficient and the value of the snow accumulation correction coefficient into the comprehensive forest fire risk prediction model to obtain the comprehensive forest fire risk value of the forest area to be assessed; Based on the comprehensive forest fire risk value of the forest area to be evaluated and the forest fire risk level classification threshold, the comprehensive forest fire risk level of the forest area to be evaluated is obtained.

2. The method according to claim 1, characterized in that: The first-level indicators of the comprehensive forest fire risk assessment index system include hazard indicators and exposure indicators. The secondary indicators of the hazard index include vegetation factors, meteorological factors, terrain factors and fire source factors. The tertiary indicators of the vegetation factors include vegetation type and EDVI double-layer vegetation index. The tertiary indicators of the meteorological factors include precipitation, snow accumulation, wind speed, temperature, relative humidity and drought level. The tertiary indicators of the terrain factors include altitude, slope and slope direction. The tertiary indicators of the fire source factors include natural fire sources and man-made fire sources. The quaternary indicators of man-made fire sources include the spatial distribution of settlements and road network density. The secondary indicators of the exposure index include population density and GDP spatial distribution.

3. The method according to claim 2, characterized in that: The forest fire risk prediction model is expressed as follows: Among them, P 综合火险 is the comprehensive forest fire risk value, C r is the precipitation correction coefficient. When the 24h precipitation is greater than or equal to 1mm, the value is 0; when the 24h precipitation is less than 1mm, the value is 1; C s is the snow accumulation correction coefficient. When the 24-hour snow depth is greater than 0, the value is 0; when the 24-hour snow depth is equal to 0, the value is 1; U FFDI W is the forest fire weather index; FFDI is the index weight of meteorological factors; A i is the quantitative value of the ith indicator, excluding meteorological factors, including 10 indicators: vegetation type, vegetation index, altitude, slope, slope aspect, natural fire source, spatial distribution of distance to residential areas, road network density, population density, and GDP spatial distribution; W i is the indicator weight of the ith indicator, excluding meteorological factors; n is the number of evaluation indicators, excluding meteorological factors; K is the fire point monitoring sensitivity. When a fire point is detected, the area where the fire point is located is directly considered to be extremely high risk, with a value of 100; if no fire point is detected, the value is 0; The forest fire risk meteorological index is expressed as follows: U FFDI =f(V)+f(T)+f(r RH )+f(M) Where V is wind speed, T is temperature, r RH is the relative humidity, M is the number of consecutive days without precipitation, f(V) is the quantitative value of the wind speed index, f(T) is the quantitative value of the temperature index, and f(r RH ) is the quantitative value of the relative humidity index, and f(M) is the quantitative value of the continuous precipitation-free day index.

4. The method according to claim 3, characterized in that: The method for obtaining the index weight of each index factor in the forest comprehensive fire risk assessment index system includes: The index weight of each index factor in the forest comprehensive fire risk assessment index system is obtained through the gray correlation method and the improved hierarchical analysis method.

5. The method according to claim 4, characterized in that: The grayscale correlation method and the improved analytic hierarchy process are used to obtain the index weights of each index factor in the forest fire risk comprehensive assessment index system, including: Obtain historical forest burnt area data and corresponding historical data of each indicator factor in the forest comprehensive fire risk assessment indicator system; The historical forest burnt area data is used as a reference sequence, and the corresponding historical data of each indicator factor in the forest comprehensive fire risk assessment index system is used as a comparison sequence; Based on the comparison sequence and the reference sequence, calculating the correlation coefficient and correlation degree of each indicator factor in the comprehensive forest fire risk assessment index system; Based on the correlation coefficient and correlation degree of each indicator factor in the forest fire risk assessment index system, the correlation order of the indicator factors in the forest fire risk assessment index system is obtained; the correlation order represents the correlation degree ranking; Based on the index level of each index factor in the forest fire risk comprehensive assessment index system, a judgment matrix is ​​constructed; the judgment matrix is ​​used to compare the importance of two index factors in the same index level; the importance of the index factor is obtained based on the correlation sequence of the index factors in the forest fire risk comprehensive assessment index system and a preset scale scoring standard; the preset scale scoring standard is a scale table of 1 to 20; Obtaining the maximum eigenvalue of the judgment matrix, and calculating a random consistency index based on the maximum eigenvalue; When the random consistency index meets the consistency test threshold, the square root method is used to calculate the maximum eigenvalue and eigenvector of the judgment matrix; and the weight of each indicator factor in the comprehensive forest fire risk assessment index system is determined based on the eigenvector.

6. The method according to claim 1, characterized in that: The step of obtaining the comprehensive forest fire risk level of the forest area to be evaluated based on the comprehensive forest fire risk value of the forest area to be evaluated and the forest fire risk level classification threshold value comprises: If P 综合火险 When the fire risk is less than 39.8, the corresponding comprehensive forest fire risk level is level one, which is a low fire risk. If 39.8≤P 综合火险 When the fire risk is less than 47.2, the corresponding comprehensive forest fire risk level is Level 2, which is a relatively low fire risk. If 47.2≤P 综合火险 When the value is less than 66.4, the corresponding comprehensive forest fire risk level is level 3, which is a relatively high fire risk; If 66.4≤P 综合火险 When the temperature is less than 73, the corresponding comprehensive forest fire risk level is level 4, which is a high fire risk; If 73≤P 综合火险 When it is less than 100, the corresponding comprehensive forest fire risk level is level 5, which is an extremely high fire risk.

7. The method according to claim 3, characterized in that: The method further comprises: In the case where fire point information is monitored in the forest area to be evaluated, determining a value of the fire point monitoring sensitivity; The value of the fire point monitoring sensitivity, the value of the precipitation correction coefficient and the value of the snow accumulation correction coefficient are input into the comprehensive forest fire risk prediction model to obtain the comprehensive forest fire risk value of the forest area to be evaluated.

8. The method according to claim 3, characterized in that: The forecast data of the natural fire source in the second future time period includes lightning warning information; The method for obtaining the quantitative value of the index of the natural fire source includes: Based on the lightning warning information and the lightning warning classification rules, a quantitative index value of the natural fire source is obtained; The step of performing quantization processing on the predicted data based on the quantization rule of the indicator factor to obtain the indicator quantization value of the indicator factor includes: Based on the classification rule of the indicator factor, determine the target category to which the predicted data of the indicator factor belongs; Based on the category quantitative scoring rule of the indicator factor, the indicator quantitative value of the indicator factor is obtained.

9. A forest fire comprehensive risk prediction device, characterized in that: include: A coefficient determination module, used to determine the value of the precipitation correction coefficient and the value of the snow accumulation correction coefficient based on the precipitation data and the snow accumulation data of the forest area to be evaluated in the first time period; A fire point acquisition module is used to determine whether fire point information is monitored in the forest area to be evaluated; An indicator acquisition module, used for acquiring the forecast data of each indicator factor in the comprehensive forest fire risk assessment indicator system of the forest area to be assessed in the future second time period when no fire point information is monitored in the forest area to be assessed; And for the predicted data of each indicator factor, the predicted data is quantized based on the quantization rule of the indicator factor to obtain the indicator quantization value of the indicator factor; A fire risk assessment module is used to input the index quantification value of each index factor in the forest comprehensive fire risk assessment index system, the value of the precipitation correction coefficient and the value of the snow accumulation correction coefficient into the forest comprehensive fire risk prediction model to obtain the forest comprehensive fire risk value of the forest area to be assessed; The level determination module is used to obtain the comprehensive forest fire risk level of the forest area to be evaluated based on the comprehensive forest fire risk value of the forest area to be evaluated and the forest fire risk level classification threshold.

10. An electronic device, characterized in that: include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory to implement the method according to any one of claims 1 to 8.

Images