CN-115527330-B - Forest fire prevention intelligent monitoring system and method based on big data

CN115527330BCN 115527330 BCN115527330 BCN 115527330BCN-115527330-B

Abstract

The invention relates to a forest fire prevention intelligent monitoring system and a method based on big data, which are based on the big data, the system comprises a monitoring headquarter, an unmanned aerial vehicle, a large database, an algorithm module, a data acquisition module, a data analysis module, a data storage module, a data processing module, a safety house and a transmission module. The invention utilizes big data technology and improves the algorithm through the temporary data packet, thereby further improving the accuracy of the algorithm for predicting whether forest fires will occur or not, and being used for precisely restraining the occurrence of fires from the source. According to the invention, the forest is divided into a plurality of areas to be monitored for monitoring, so that the forest fireproof information is effectively mastered. In addition, the safety house is further attached to life practice by building, the feasibility is good, the information transmission effect is obviously improved, and the safety house is particularly significant in the aspect of dealing with small-scale fire and guaranteeing the safety of trapped people.

Inventors

CHU SHIWEI
LI FEI
YUAN SHIHUI
ZHAO FEIFEI
Ding Huaibao
YE BOYANG
WANG JUNFEI
XU LIBO
JING XIAOBING
CUI XIN
CHENG GUOXU
ZHANG WEI

Assignees

天立泰科技股份有限公司

Dates

Publication Date: 20260512
Application Date: 20220919

Claims (9)

1. The forest fire prevention intelligent monitoring system based on big data is characterized by comprising a monitoring headquarter, an unmanned plane, a big database, an algorithm module, a data acquisition module, a data analysis module, a data storage module, a data processing module, a safety house and a transmission module, The unmanned aerial vehicle is used for inspecting forests; The large database stores all historical data of forests; The algorithm module is used for operation; The data acquisition module is used for acquiring data; the data analysis module is used for analyzing data; The data storage module is used for storing data; the data processing module is used for processing data; the transmission module is used for transmitting data; The method comprises the steps of dividing a forest into a plurality of areas to be monitored, calling out historical data of the areas to be monitored from a large database, and obtaining forest fire influence factors of each area to be monitored through a data processing module, constructing an initial prediction function F 1 in an algorithm module, collecting real-time data through a data collection module, and performing deep learning and adjustment on the initial prediction function F 1 in the algorithm module according to the historical data in the large database, the real-time data collected in real time and a temporary data packet to obtain a new prediction function F 2 ; Transmitting the real-time data acquired in real time, the temporary data packet and the output result of the new prediction function F 2 to a monitoring headquarter through a transmission module, and storing the real-time data, the temporary data packet and the output result at a data storage module; The monitoring headquarter analyzes through the data analysis module, and determines whether to dispatch unmanned aerial vehicles and forest patrols according to analysis results; the fire-proof liquid storage box (10) is buried below the house bottom of the safety house, a decorative layer (1), a heat-insulating fire-proof layer (2), a reinforced concrete layer (3) and a fire-proof layer (4) are sequentially arranged on the wall of the safety house from inside to outside, the heat-insulating fire-proof layer (2) is made of a fire-proof rock wool board, the fire-proof layer (4) is made of a fire-proof board, steel bars embedded in the reinforced concrete layer (3) extend to the inside of the fire-proof liquid storage box (10), a plurality of bundle-shaped capillary tube sets (20) are embedded in the reinforced concrete layer (3), and the upper ends of the capillary tube sets (20) are embedded in the inside of the reinforced concrete layer (3), and the lower ends of the capillary tube sets (20) are arranged in the fire-proof liquid storage box (10); The capillary collection (20) comprises a round tubular porous metal sleeve (22) and seven capillary bundles (21) arranged inside the porous metal sleeve (22), the porous metal sleeve (22) is a metal pipe body with a plurality of round holes distributed on the surface, the capillary bundles (21) comprise porous metal pipes (213), six porous capillaries (212) arranged inside the porous metal pipes (213) and reinforced capillaries (211) coaxially arranged with the porous metal pipes (213), the six porous capillaries (212) surround the outside of the reinforced capillaries (211), the six porous capillaries (212) are twisted, the porous capillaries (212) are made of stainless steel capillaries, a plurality of through holes are drilled on the surface of the stainless steel capillaries by adopting a laser drilling technology, the reinforced capillaries (211) are made of stainless steel capillaries on the periphery of the stainless steel capillaries in a pressurizing mode, an included angle between any two side surfaces (2111) is 60 degrees, a triangular area (23) is formed by the inner wall of the porous metal sleeve (22) and an area between two adjacent capillary bundles (21), and the porous capillaries (212) are filled with a triangular area (23), and the number of twisted porous capillaries (212) is 3-4.
2. The forest fire prevention intelligent monitoring system based on big data, as set forth in claim 1, wherein the capillary tube collection (20) sequentially comprises a cylindrical section (20 a), a round-down table section (20 b) and a sharp section (20 c) with a downward tip from top to bottom, the cylindrical section (20 a) is an upper section of a porous metal sleeve (22), the round-down table section (20 b) is a lower section of the porous metal sleeve (22), the sharp section (20 c) is composed of a lower end of a porous capillary tube (212) and a lower end of a reinforced capillary tube (211), a packaging mechanism (30) is installed at the lower part of the capillary tube collection (20), the packaging mechanism (30) is arranged inside a fire prevention liquid storage tank (10), the packaging mechanism (30) comprises a sealing sleeve (31) sleeved outside the cylindrical section (20 a), the upper end of the sealing sleeve (31) is in sealing connection with the outer wall of the cylindrical section (20 a), the round-down table section (20 b) and the sharp section (20 c) are both arranged inside the sealing sleeve (22), the round-down table section (20 b) and the sharp section (20 c) are arranged at the inner end of the sealing sleeve (31), the round-down table section (35) is arranged at the lower end of the round-down table section (35) and the round-down table section is provided with a sealing plate (35), the inner ring of the metal plate body (35) is connected with a film layer (34) in a sealing way, a first graphite layer (33) is connected between the periphery of the metal plate body (35) and the inner wall of the sealing sleeve (31) in a sealing way, a check ring (32) is arranged below the metal plate body (35), and the check ring (32) is fixedly connected with the lower end of the sealing sleeve (31).
3. The forest fire prevention intelligent monitoring system based on big data, as set forth in claim 2, wherein the bottom center of the fire prevention liquid storage tank (10) is provided with an upward protruding and volcano-shaped protruding part (12), a driving mechanism (50) for breaking a film layer (34) is arranged between the top of the fire prevention liquid storage tank (10) and the protruding part (12), an annular groove (13) is formed between the protruding part (12) and the side wall of the fire prevention liquid storage tank (10), the packaging mechanism (30) is arranged along the annular groove (13), a storage cavity (11) for storing fire prevention liquid is formed between the outer side of the driving mechanism (50) and the inner wall of the fire prevention liquid storage tank (10), the driving mechanism (50) comprises a film cylinder (52), a film coated paper cylinder (51) arranged in the center of the film cylinder (52), a slide rod (54) coaxially arranged with the film coated paper cylinder (51), a pressing plate (510) fixedly arranged at the upper end of the slide rod (54), the upper end of the film coated paper cylinder (51) is connected with the inner wall of the fire prevention liquid storage tank (10), the film coated paper cylinder (51) is connected with the inner wall (52) of the annular paper cylinder (52), the annular cavity (53) is internally stored with perfluorinated hexanone; the top of fireproof liquid storage box (10) is provided with the perforation that supplies slide bar (54) to slide, the upper end and the clamp plate (510) of slide bar (54) all set up in the top of fireproof liquid storage box (10), the lower extreme setting of slide bar (54) is in the inside of tectorial membrane fiber container (51), be provided with elastic ball (55) between the bottom of lower extreme of slide bar (54) and tectorial membrane fiber container (51), elastic ball (55) and the bottom fixed connection of tectorial membrane fiber container (51), the upper portion fixed mounting of slide bar (54) has stop ring (58), stop ring (58) set up the below at fireproof liquid storage box (10) top, the outside of slide bar (54) still overlaps and is equipped with polylith porous metal sheet (56), porous metal sheet (56) and slide bar (54) fixed connection, the inside of tectorial membrane fiber container (51) still is provided with initiator storage area (57), initiator storage area (57) store, be connected with initiator between perforation protection casing (54) and slide bar (40), install in the top of sealing plate (40) on the top of fireproof liquid storage box (40).
4. The forest fire prevention intelligent monitoring system based on big data as set forth in claim 3, wherein the fire prevention liquid is prepared by mixing water, perfluoro-hexanone and tetrol according to a mass ratio of 100 (50-60) to 6-7.
5. The forest fire prevention intelligent monitoring system based on big data as set forth in claim 3, wherein the initiating agent is potassium chlorate, red phosphorus and potassium bicarbonate in a mass ratio of (3.3-3.5) to (1:31).
6. A forest fire prevention intelligent monitoring method based on big data is characterized in that, the forest fire prevention intelligent monitoring system based on big data as set forth in claim 1 is adopted, comprising the following steps: Step S1, dividing a forest into a plurality of areas to be monitored, calling out historical data of the areas to be monitored from a large database, and obtaining a forest fire influence factor of each area to be monitored; Step S2, performing deep learning and adjustment on the initial prediction function F 1 according to historical data in a large database, real-time data acquired in real time and a temporary data packet to obtain a new prediction function F 2 , a new prediction function F 2 , wherein an output result of the new prediction function F 2 comprises prompt information and alarm information; when the output is prompt information, the corresponding input value is input X nt , the prompt information is transmitted to a safety house through a transmission module, and meanwhile forest patrolling personnel are dispatched to be checked and fed back in the field; When the output is alarm information, the corresponding input value is input X ng , and meanwhile, the unmanned plane and a forest inspector are dispatched to be checked and fed back in the field; S3, marking the prompt information and the input X nt when the prompt information is verified to be 'wrong' after being checked in the field, and incorporating the prompt information and the input X nt into a temporary data packet; When the alarm information is checked in the field and verified to be "wrong", marking the alarm information and the input X nt , and incorporating the alarm information and the input X nt into a temporary data packet.
7. The intelligent forest fire prevention monitoring method based on big data as set forth in claim 6, wherein the forest fire influence factors comprise climate factors, geographical factors, woodland factors, seasonal factors and artificial factors, the climate factors comprise temperature, humidity, drought frequency and precipitation, the geographical factors comprise landforms, terrains and soil, the woodland factors comprise plant number density, canopy density, accumulation amount and forestation density, the seasonal factors comprise highest air temperature in different seasons and fire conditions in different solar terms, and the artificial factors comprise flow of people entering the forest and artificial fire conditions.
8. The intelligent forest fire prevention monitoring method based on big data as set forth in claim 6, wherein the forest fire influence factor is obtained by three modes of algorithm grabbing, manual input and custom setting.
9. The intelligent forest fire prevention monitoring method based on big data as set forth in claim 6, wherein the safety house is set according to the landform features and the cruising mileage of the unmanned aerial vehicle, the safety house is provided with an information receiving module for receiving prompt information and alarm information, a rest area for personnel to rest, a storage area, a reporting area, an unmanned aerial vehicle stay and charging area, The personnel comprise forest patrols, forest firefighters, overhaulers and forest policemen, living materials and fire-fighting materials are stored in the storage area, communication content can be sent to a monitoring headquarter in the reporting area, and the unmanned aerial vehicle stays in the unmanned aerial vehicle stay and charging area temporarily and charges.

Description

Forest fire prevention intelligent monitoring system and method based on big data Technical Field The invention relates to a forest fire prevention intelligent monitoring system and method based on big data, and belongs to the technical field of forest intelligent fire prevention. Background The conditions required for the occurrence of the fire disaster are that the forest fire disaster must have combustible matters, fire weather and fire sources, one of the three conditions is lacking, and the forest fire disaster cannot occur. Through scientific calculation, the fire is caused by human reasons to be more than 95%, so a great deal of facts indicate that forest fires can be prevented, for example, flammable matters and fire sources are strictly controlled manually, and fire hazard weather is predicted and forecasted, so that the occurrence of the fire can be greatly reduced. Currently, forest fire risk level prediction is mainly based on stacking algorithm. However, the method designs a processing technology of massive space-time data, realizes modeling driven by data, and predicts forest fires based on the model. The technology only collects a large amount of data for each factor of partial factors to carry out modeling analysis, for example, factors such as combustibles, fire weather, fire sources and the like, and other important factors (such as geographic factors, woodland factors, artificial factors and the like) are ignored, so that the algorithm has obvious one-sided performance. In the prior art, most of the existing technical means are emergency solving methods and post-treatment work of forest fires. And control from the forest fire source to reach the purpose of preventing the fire, lack corresponding technical means again, predictive mode exists limitation and deficiency, causes the result inaccurate. Once a forest fire occurs, the loss caused by the forest fire is not estimated, so that a forest fire intelligent monitoring system and method based on big data are needed for precisely suppressing the occurrence of the fire from the source. Disclosure of Invention Aiming at the defects existing in the prior art, the invention provides a forest fire prevention intelligent monitoring system and method based on big data, and the specific technical scheme is as follows: a forest fire prevention intelligent monitoring method based on big data comprises the following steps: Step S1, dividing a forest into a plurality of areas to be monitored, calling out historical data of the areas to be monitored from a large database, and obtaining a forest fire influence factor of each area to be monitored; Step S2, performing deep learning and adjustment on the initial prediction function F 1 according to historical data in a large database, real-time data acquired in real time and a temporary data packet to obtain a new prediction function F 2, a new prediction function F 2, wherein an output result of the new prediction function F 2 comprises prompt information and alarm information; when the output is prompt information, the corresponding input value is input X nt, the prompt information is transmitted to a safety house through a transmission module, and meanwhile forest patrolling personnel are dispatched to be checked and fed back in the field; When the output is alarm information, the corresponding input value is input X ng, and meanwhile, the unmanned plane and a forest inspector are dispatched to be checked and fed back in the field; S3, marking the prompt information and the input X nt when the prompt information is verified to be 'wrong' after being checked in the field, and incorporating the prompt information and the input X nt into a temporary data packet; When the alarm information is checked in the field and verified to be "wrong", marking the alarm information and the input X nt, and incorporating the alarm information and the input X nt into a temporary data packet. According to the technical scheme, the forest fire influence factors comprise climate factors, geographic factors, woodland factors, seasonal factors and artificial factors, the climate factors comprise temperature, humidity, drought frequency and precipitation, the geographic factors comprise landforms, terrains and soil, the woodland factors comprise plant number density, canopy density, accumulation amount and forestation density, the seasonal factors comprise highest air temperature in different seasons and fire conditions in different solar terms, and the artificial factors comprise people flow entering forests and artificial fire conditions. According to the technical scheme, the forest fire influence factors are obtained by three modes of algorithm grabbing, manual input and custom setting. According to the technical proposal, the safety house is arranged according to the landform characteristics and the cruising mileage of the unmanned aerial vehicle, the safety house is provided with an information receiving module for receiv