Meteorological-aided early warning method for forest fire danger

Abstract

The invention discloses a weather-assisted forest fire early warning method, which comprises the following steps of: constructing a weather early warning model, calculating an estimated value of a model coefficient through historical data of fire occurrence and corresponding weather historical data, and substituting the estimated value into the weather early warning model to enable the weather early warning model to become a computable weather early warning model. In combination with the computable meteorological early warning model and meteorological data obtained through meteorological monitoring equipment or weather forecast, a meteorological risk index can be obtained so as to further realize early warning of forest fire danger. According to the weather-assisted forest fire early warning method, the early warning identification efficiency and accuracy of the forest fire can be improved, hidden fire sources can be effectively identified, and the problems of waste of fire protection resources and the like caused by false alarm are avoided.

Description

Technical Field

[0001] The present invention relates to the technical field of forest fire risk early warning, and in particular to a meteorologically assisted forest fire risk early warning method. Background Art

[0002] The management and prevention of forest fire risk is a global challenge, especially in the context of climate warming and increased human activities. Forest fires not only damage the ecosystem, but also threaten human life and property. Therefore, an effective forest fire early warning system is essential to reduce the occurrence of fires and control the spread of fires.

[0003] The occurrence of forest fire risk is usually affected by three main factors, namely, ignition sources (natural phenomena such as lightning or human activities such as smoking, barbecue, campfire, etc.), combustible materials (tree species, fallen leaves, etc.) and burning conditions (temperature, humidity, etc.).

[0004] At present, forest fire risk warning mainly relies on post-event monitoring methods, that is, after the fire occurs, the fire source is discovered through monitoring means, and then early warning and prevention and control are carried out. Commonly used monitoring methods include high-point camera monitoring, satellite fire source hotspot analysis and drone reconnaissance.

[0005] However, the existing forest fire early warning methods have some defects: on the one hand, there is a certain time lag in discovering the fire through the fire point, which may cause the fire to spread and miss the best time to extinguish the fire. On the other hand, for some hidden fire sources, they can only be identified by the early warning system when the fire is large, which is easy to miss. In addition, the existing forest fire early warning methods may have false alarms, and incorrect early warnings can easily lead to a waste of firefighting resources. Summary of the invention

[0006] The purpose of the present invention is to solve the technical problems in the above-mentioned existing forest fire risk warning methods, such as time lag, difficulty in identifying hidden fire sources, and easy false alarms leading to waste of fire-fighting resources, and to provide a meteorological-assisted forest fire risk warning method that can improve the efficiency and accuracy of forest fire risk warning identification.

[0007] In order to solve the above problems, the present invention is implemented according to the following technical solutions:

[0008] The present invention provides a meteorologically assisted forest fire risk early warning method, the method comprising:

[0009] S100: Establish a meteorological warning model. The meteorological warning model is

[0010]

[0011] Wherein, r is the meteorological risk index of forest fire in the selected spatial area, H is the forest fire risk index of relative humidity, M is the forest fire risk index of temperature, F is the forest fire risk index of wind speed, x is latitude, y is longitude, z is altitude, t is time; n is the values ​​of the three meteorological elements of relative humidity, temperature and wind speed monitored or forecasted in the selected spatial area within a certain period of time; α, β, δ, θ 1 ,θ 2 ,θ 3 and θ 4 All are calculation coefficients;

[0012] S200, based on the historical data of fire risk and the corresponding historical meteorological data, calculating an estimated value of a calculation coefficient of a meteorological warning model by using one of an artificial intelligence model or a regression analysis model;

[0013] S300, substituting the estimated value into the meteorological warning model to obtain an estimation model of the meteorological warning model;

[0014] S400: In the selected spatial region, the meteorological observation data or the meteorological forecast data is substituted into the estimation model to obtain the meteorological risk index.

[0015] S500, when the meteorological risk index Less than or equal to the preset minimum meteorological risk index value r 1 At this time, the meteorological risk index warning level is low risk.

[0016] Preferably, S210, based on the historical data of fire risk and the corresponding historical meteorological data, the forest fire risk index H of relative humidity, the forest fire risk index M of temperature and the forest fire risk index F of wind speed are calculated.

[0017] Preferably, S211, according to the formula The forest fire risk index H of relative humidity is calculated, wherein the RH is a relative humidity value obtained based on historical data of fire risk and corresponding historical meteorological data.

[0018] Preferably, S212, according to the formula M = (lnT) 2 The forest fire risk index M of the temperature is calculated, wherein T is the temperature value obtained based on the historical data of the fire risk and the corresponding historical meteorological data.

[0019] Preferably, S213, the forest fire risk index F of wind speed is calculated according to the formula F=lnW, wherein F is a wind speed value obtained based on historical data of fire risks and corresponding meteorological historical data.

[0020] Preferably, the value range of RH is 1%-100%, when the acquired RH value is not within the value range of RH, the value of the forest fire risk index H is 0; the value range of T is 223.15K-323.15K, when the acquired T value is not within the value range of T, the value of the forest fire risk index M is 0; the value range of W is 1 m / s-30 m / s, when the acquired W value is not within the value range of W, the value of the forest fire risk index F is 0.

[0021] Preferably, S220, set r=0 when there is no forest fire risk, and set r=1 when there is a forest fire risk; S230, substitute the forest fire risk index H of relative humidity, the forest fire risk index M of temperature and the forest fire risk index F of wind speed calculated in step S210 into the meteorological warning model, and based on the historical data of fire risks and the corresponding meteorological historical data, calculate the estimated value of the calculation coefficient of the meteorological warning model through one of the artificial intelligence model and the regression analysis model.

[0022] Preferably, when the meteorological risk index Greater than or equal to the preset maximum meteorological risk index value r h When the meteorological risk index warning level is high risk; when the meteorological risk index Greater than the preset minimum meteorological risk index value r 1 And less than the preset maximum meteorological risk index value r h At this time, the meteorological risk index warning level is medium risk.

[0023] Compared with the prior art, the present invention has the following beneficial effects:

[0024] The present invention provides a meteorologically assisted forest fire risk early warning method, by constructing a meteorological early warning model, and calculating the estimated value of the model coefficient through the historical data of the fire risk and the corresponding meteorological historical data, substituting the estimated value into the meteorological early warning model to make it a computable meteorological early warning model, combining the computable meteorological early warning model and the meteorological data obtained through meteorological monitoring equipment or weather forecast, it is possible to obtain a meteorological risk index to further realize the early warning of forest fire risk. Through the meteorologically assisted forest fire risk early warning method of the present invention, the early warning recognition efficiency and accuracy of forest fire risk can be improved, and hidden fire sources can be effectively identified, avoiding the problem of waste of fire fighting resources due to false alarms. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] The specific embodiments of the present invention are further described in detail below in conjunction with the accompanying drawings, wherein:

[0026] Figure 1 The present invention is a structural principle diagram of a meteorological-assisted forest fire risk early warning method. DETAILED DESCRIPTION

[0027] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0028] The terms used in this application are only for the purpose of describing specific embodiments and are not intended to limit this application. Unless otherwise defined, the technical terms or scientific terms used in this specification should be understood by people with ordinary skills in the field to which this application belongs. The words "first", "second" and similar words used in this specification and claims do not indicate any order, quantity or importance, but are only used to distinguish different technical features.

[0029] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0030] Figure 1 The present invention is a structural principle diagram of a meteorological-assisted forest fire risk early warning method.

[0031] like Figure 1 As shown, the present invention provides a meteorologically assisted forest fire risk early warning method, the method comprising the following execution steps:

[0032] S100, establish a meteorological warning model, the meteorological warning model is

[0033]

[0034] Among them, r is the meteorological risk index of forest fire in the selected spatial area, H is the forest fire risk index of relative humidity, M is the forest fire risk index of temperature, F is the forest fire risk index of wind speed, x is latitude, y is longitude, z is altitude, t is time; n is the values ​​of relative humidity, temperature and wind speed monitored or forecasted in the selected spatial area within a certain period of time; α, β, δ, θ 1 ,θ 2 ,θ 3 and θ 4 All are calculation coefficients.

[0035] It should be noted that meteorological conditions are an important factor affecting the forest fire risk level because they directly affect the dryness of combustibles and the possibility of fire spread. Among them, relative humidity is the ratio of the water vapor content in the air to the maximum water vapor content that the air can accommodate at that temperature. The lower the relative humidity, the drier the air, and the lower the moisture content of combustibles (such as leaves, dead branches, etc.), which are more likely to be ignited. Under low humidity conditions, vegetation in the forest is more likely to lose moisture, increasing the risk of fire. The decrease in relative humidity will also accelerate the evaporation of moisture in the soil and vegetation, further reducing the humidity of the soil and vegetation and increasing the fire risk level. Rising temperatures will increase the dryness of combustibles in the forest, making them more likely to be ignited. High temperatures will also accelerate the evaporation of moisture in the soil and vegetation, further reducing humidity and increasing the fire risk level. Rising temperatures may also be accompanied by heat waves, which will make the environment in the forest drier and increase the fire risk level. In addition, an increase in wind speed can accelerate the spread of a fire. Wind can not only blow the fire source to a new area, but also make the fire more intense. An increase in wind speed will also reduce relative humidity, because the wind can take away moisture from the ground, making the air drier. Strong winds may also blow up combustibles on the ground, forming flying fires, increasing the risk of fire spreading. Therefore, by monitoring meteorological conditions such as relative humidity, temperature and wind speed, the changing trend of forest fire risk levels can be predicted more accurately and efficiently. For example, when relative humidity decreases, temperature rises and wind speed increases, the forest fire risk level may rise rapidly. Therefore, by real-time monitoring of these meteorological factors, possible fire risk situations can be warned in advance, so that corresponding preventive measures can be taken, such as limiting fire sources, strengthening patrols, and deploying firefighting resources in advance, so as to reduce the occurrence and spread of fires.

[0036] S200. Based on the historical data of fire risks and the corresponding historical meteorological data, an estimated value of a calculation coefficient of a meteorological warning model is calculated by using one of an artificial intelligence model and a regression analysis model.

[0037] It should be noted that by calculating the estimated value of the calculation coefficient of the meteorological warning model through one of the artificial intelligence models or the regression analysis model, this method can take into account the differences in the risk of fire occurrence in different regions and time periods, making the warning model more flexible and accurate, and improving the efficiency and accuracy of forest fire risk warning identification.

[0038] It can be understood that, on the one hand, by analyzing the relationship between historical fire events and meteorological conditions, it is possible to identify which meteorological factors have a significant impact on the occurrence of fire risk. For example, meteorological factors such as temperature, humidity, wind speed and precipitation are closely related to the occurrence of forest fire risk. Through historical data analysis, the strength of the correlation between these meteorological factors and the occurrence of fire risk can be determined, thereby providing important input parameters for the model. On the other hand, using historical data to estimate the calculation coefficients of the model can improve the accuracy of the early warning model. This is because historical data contains the actual situation when fire risk events occurred in the past, and the model parameters calculated from these data can better reflect the actual situation, thereby improving the prediction ability of the model. In addition, historical data is not only used to establish the model, but also to verify the prediction effect of the model. By comparing the model prediction results with the actual fire risk events, the accuracy of the model can be evaluated, and the model can be adjusted and optimized as needed to improve its prediction ability.

[0039] Furthermore, selecting an appropriate historical data range is crucial for building a meteorological warning model, because historical data can provide the model with contextual information about past events and help the model learn and predict future trends. The following are several key factors that can be used to determine an appropriate historical data range: (1) Data representativeness: The historical data selected should be representative of current and future conditions. For example, if climate conditions have changed significantly over time, more recent data may be more representative. (2) Data completeness: The dataset should be long enough to capture the natural variability of meteorological conditions, including seasonal and interannual changes. This helps the model identify and learn fire risk trends under different conditions. (3) Data quality: The quality of historical data is critical to the accuracy of the model. The data should be quality controlled to ensure its accuracy and reliability. (4) Relevance: Select meteorological factors that are highly correlated with forest fire risk, such as temperature, humidity, wind speed, and precipitation. These factors have a direct impact on the occurrence and development of fire risk. (5) Statistical significance: The data range selected should be long enough to allow statistical analysis and ensure the significance of the results. For example, multiple years of data can be used to analyze the changing trend of the fire risk meteorological index. (6) Availability and accessibility: The selected data should be available and can be used for model building. Many meteorological data can be obtained from sources such as the National Meteorological Information Center and the Energy Big Data Platform. (7) Purpose of the model: Consider the ultimate use of the model. If the model is intended to predict fire risk in a specific season, then the selected data should cover those seasons. If the model is used for long-term trend analysis, a longer-term data set is required. (8) Regional specificity: The climatic conditions and forest types in different regions vary greatly, so the selected data should reflect the specific conditions of these regions. For example, the distribution characteristics of the fire prevention period and fire risk meteorological index in the two major forest regions of Northeast and Southwest are different. (9) Historical fire events: If possible, include data on major fire events that have occurred in history, which helps the model learn the meteorological conditions when the fire occurred. (10) Update frequency: The update frequency of the data should match the prediction frequency of the model. For example, if the model requires real-time prediction, then the data should also be updated in real time. Therefore, considering the above factors and the actual application situation, an appropriate historical data range can be selected to build an effective meteorological warning model.

[0040] S210. Based on the historical data of fire risks and the corresponding historical meteorological data, calculate the forest fire risk index H of relative humidity, the forest fire risk index M of temperature, and the forest fire risk index F of wind speed.

[0041] S211, according to the formula The forest fire risk index H of relative humidity is calculated, where RH is the relative humidity value obtained based on the historical data of fire risk and the corresponding historical meteorological data.

[0042] Furthermore, the value range of RH is 1%-100%. When the acquired RH value is not within the value range of RH, the value of the forest fire risk index H is 0.

[0043] S212, according to the formula M = (lnT) 2 Calculate the forest fire risk index M of the temperature, where T is the temperature value obtained based on the historical data of fire risks and the corresponding historical meteorological data.

[0044] Furthermore, the value range of T is 223.15K-323.15K. When the obtained value of T is not within the value range of T, the value of the forest fire risk index M is 0.

[0045] S213. Calculate the forest fire risk index F of wind speed according to the formula F=lnW, where F is the wind speed value obtained based on the historical data of fire risk and the corresponding historical meteorological data.

[0046] Furthermore, the value range of W is 1 m / s-30 m / s. When the obtained value of W is not within the value range of W, the value of the forest fire risk index F is 0.

[0047] S220, setting r=0 when there is no forest fire risk, and r=1 when there is a forest fire risk;

[0048] S230. Substitute the forest fire risk index H of relative humidity, the forest fire risk index M of temperature and the forest fire risk index F of wind speed calculated in step S210 into the meteorological warning model, and calculate the estimated value of the calculation coefficient of the meteorological warning model through one of the artificial intelligence model and the regression analysis model based on the historical data of fire risks and the corresponding historical meteorological data.

[0049] It can be understood that, from the historical data of fire risk and the corresponding historical meteorological data, the latitude x, longitude y, altitude z, time t and the values ​​n of the three meteorological elements of relative humidity, temperature and wind speed monitored or forecasted in the selected spatial area within a certain time period can be obtained. Combined with the forest fire risk index H of relative humidity, the forest fire risk index M of temperature and the forest fire risk index F of wind speed calculated in step S210, the estimated value of the calculation coefficient of the meteorological warning model can be calculated by one of the artificial intelligence model or the regression analysis model, which can be recorded as

[0050] S300: Substitute the estimated value into a meteorological warning model to obtain an estimation model of the meteorological warning model.

[0051] S400: In the selected spatial area, the meteorological observation data or meteorological forecast data is substituted into the estimation model to obtain the meteorological risk index.

[0052] It can be understood that from the meteorological observation data or meteorological forecast data, relative humidity RH, temperature T, wind speed W, latitude x, longitude y, altitude z, time t and the values ​​n of the three meteorological elements of relative humidity, temperature and wind speed monitored or forecasted in the selected spatial area within a certain period of time can be obtained, and the forest fire risk index H of relative humidity, the forest fire risk index M of temperature and the forest fire risk index F of wind speed can be further calculated, thereby further obtaining the meteorological risk index meteorological risk index

[0053] S500, current meteorological risk index Less than or equal to the preset minimum meteorological risk index value r 1 At this time, the meteorological risk index warning level is low risk.

[0054] In this embodiment, when the meteorological risk index Greater than or equal to the preset maximum meteorological risk index value r h When the meteorological risk index warning level is high risk; when the meteorological risk index Greater than the preset minimum meteorological risk index value r 1 And less than the preset maximum meteorological risk index value r h At this time, the meteorological risk index warning level is medium risk.

[0055] In summary, the meteorologically assisted forest fire risk early warning method described in the present invention constructs a meteorological early warning model, and calculates the estimated value of the model coefficient through the historical data of the fire risk and the corresponding meteorological historical data, and substitutes the estimated value into the meteorological early warning model to make it a computable meteorological early warning model. In combination with the computable meteorological early warning model and the meteorological data obtained by the meteorological monitoring equipment or the weather forecast, the meteorological risk index can be obtained to further realize the early warning of the forest fire risk. Through the meteorologically assisted forest fire risk early warning method of the present invention, the early warning recognition efficiency and accuracy of the forest fire risk can be improved, and the hidden fire source can be effectively identified, avoiding the problem of waste of fire fighting resources due to false alarms.

[0056] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Therefore, any modification, equivalent change and modification made to the above embodiment according to the technical essence of the present invention without departing from the technical solution of the present invention shall still fall within the scope of the technical solution of the present invention.

Claims

1. A meteorologically assisted forest fire risk early warning method, characterized in that: The method comprises: S100: Establish a meteorological warning model. The meteorological warning model is Wherein, r is the meteorological risk index of forest fire risk in the selected spatial area, H is the forest fire risk index of relative humidity, M is the forest fire risk index of temperature, F is the forest fire risk index of wind speed, x is latitude, y is longitude, z is altitude, and t is time; n is the values ​​of the three meteorological elements of relative humidity, temperature and wind speed monitored or forecasted in the selected spatial area within a certain time period; α, β, δ, θ1, θ2, θ3 and θ4 are all calculation coefficients; S200, based on the historical data of fire risk and the corresponding historical meteorological data, calculating an estimated value of a calculation coefficient of a meteorological warning model by using one of an artificial intelligence model or a regression analysis model; S300, substituting the estimated value into the meteorological warning model to obtain an estimation model of the meteorological warning model; S400: In the selected spatial region, the meteorological observation data or the meteorological forecast data is substituted into the estimation model to obtain the meteorological risk index. S500, when the meteorological risk index When it is less than or equal to the preset minimum meteorological risk index value r1, the meteorological risk index warning level is low risk.

2. A meteorologically assisted forest fire risk early warning method according to claim 1, characterized in that: S210. Based on the historical data of fire risks and the corresponding historical meteorological data, the forest fire risk index H of relative humidity, the forest fire risk index M of temperature and the forest fire risk index F of wind speed are calculated.

3. A meteorological-assisted forest fire risk early warning method according to claim 2, characterized in that: S211, according to the formula The forest fire risk index H of relative humidity is calculated, wherein the RH is a relative humidity value obtained based on historical data of fire risk and corresponding historical meteorological data.

4. A meteorologically assisted forest fire risk early warning method according to claim 3, characterized in that: S212, according to the formula M = (lnT) 2 The forest fire risk index M of the temperature is calculated, wherein T is the temperature value obtained based on the historical data of the fire risk and the corresponding historical meteorological data.

5. A meteorological-assisted forest fire risk early warning method according to claim 4, characterized in that: S213. Calculate the forest fire risk index F of wind speed according to the formula F=lnW, wherein F is a wind speed value obtained based on historical data of fire risks and corresponding historical meteorological data.

6. A meteorologically assisted forest fire risk early warning method according to claim 5, characterized in that: The value range of RH is 1%-100%. When the acquired RH value is not within the value range of RH, the value of the forest fire risk index H is 0; The value range of T is 223.15K-323.15K. When the acquired value of T is not within the value range of T, the value of the forest fire risk index M is 0; The value range of W is 1 m / s-30 m / s. When the acquired value of W is not within the value range of W, the value of the forest fire risk index F is 0.

7. A meteorological-assisted forest fire risk early warning method according to claim 2, characterized in that: S220, setting r=0 when there is no forest fire risk, and r=1 when there is a forest fire risk; S230. Substitute the forest fire risk index H of relative humidity, the forest fire risk index M of temperature and the forest fire risk index F of wind speed calculated in step S210 into the meteorological warning model, and calculate the estimated value of the calculation coefficient of the meteorological warning model through one of the artificial intelligence model and the regression analysis model based on the historical data of fire risks and the corresponding historical meteorological data.

8. A meteorologically assisted forest fire risk early warning method according to claim 1, characterized in that: When the meteorological risk index Greater than or equal to the preset maximum meteorological risk index value r h When the weather risk index warning level is high risk; When the meteorological risk index Greater than the preset minimum meteorological risk index value r1 and less than the preset maximum meteorological risk index value r h At this time, the meteorological risk index warning level is medium risk.

Images