KR102322427B1 - Bigdata based building fire prevention response system and method - Google Patents

Abstract

The present invention relates to a big data-based building fire prevention response system, and by monitoring the operation status of the firefighting facilities through double sensing that receives status information and sound field information in real time from firefighting facilities provided in a building to be monitored. a monitoring server that generates monitoring information; Extract feature data for real fire determination from monitoring information of firefighting facilities, put feature data as input data of neural network, classify fire information and non-fire report through machine learning, and analyze deterioration of firefighting facilities through feature data analysis an analysis server for generating information, replacement time analysis information, and non-fire information analysis information; and a management server that transmits the fire information to the associated fire department, indexes the 3D scanning space modeling data of the building in which the fire occurs from the database, and transmits it to the firefighter terminal of the competent fire station.

Description

Big data based building fire prevention response system and method}

The present invention relates to a big data-based building fire prevention response system, and more particularly, by configuring a sound field sensor-based CCTV, a flame image detector, and a double-sensing structure detector for non-fire warning prevention, fire fighting in a building to be monitored By monitoring the operation status of the facility in real time to secure the reliability of the firefighting facility, it is possible to prevent and respond to building fires through regular management of whether or not they operate normally. It relates to a technology that enables firefighters to recognize spatial information in advance and take immediate and rapid fire suppression and relief measures.

This patent was carried out with research funding support from the Ministry of Land, Infrastructure and Transport's Urban Architecture Research Project (Project No. 19AUDP-B100356-05)

With the development of science and technology, the field of firefighting facilities is also becoming more advanced, diversified, and complex, and it leads to the need for firefighting facilities that can recognize various fire situations. In such a case, there are situations in which normal operation is not performed. In Korea, when a non-fire report occurs, it is highly likely that property damage will further increase as it leads to system malfunction and delayed reporting despite the actual fire situation.

The issue of non-fire alarms continues to occur not only in Korea but also abroad. In the United States, there were 16 non-fire alarms for every 10 actual fire situations in 2009, and statistics show that 45 non-fire alarms occur per 10 structural fires. there is This leads to a waste of firefighting power, and in overseas countries, fines are imposed on the building owner through institutional devices, and fundamental inspection and management of firefighting facilities is required.

In Korea, installation targets are stipulated in accordance with the Act on Installation, Maintenance and Safety Management of Firefighting Facilities, but there is no system in place to effectively manage non-fire alarms. Therefore, it is required to establish an inspection DB and management system (software) so that non-fire alarm inspection and management can be continuously performed.

The disaster management system, the disaster prevention system, consists of some subsystems of the overall management system of the building. operation), power/lighting equipment (emergency lighting, emergency power), and access control (door opening and closing) are planned to operate according to the detection signal in case of a disaster.

In the current disaster management system, communication is limited to the extent that a signal for detecting a disaster/disaster occurrence is notified to other facilities, and the facilities notified of the disaster are performing emergency operation planned for each facility. In other words, rather than integrated/connected operation according to the occurrence status of disasters or evacuations, it is built as a 1:1 individual response.

On the other hand, emergency operation methods for each facility of the disaster/disaster management system are suggested by relevant standards such as fire safety standards, but there are no specific standards for emergency operation methods for many other facilities. is currently dependent on

Currently, ICT technology centered on R-type receivers is evaluated as the most backward and closed in the field of building equipment, and the openness of the disaster/disaster management system is declining due to the passivity centered on laws and the technology selfishness of the disaster/disaster management system provider. Field application of new technologies developed by national R&D, etc. is insufficient.

In addition, in the building sector, efforts to develop contents or solutions for complex disaster management by operating communication infrastructure are still insufficient.

For building fire safety management, local information of the surrounding area as well as detailed basic information at the level of the absence unit are required. It is difficult to recognize the exact fire situation due to inaccuracy and the complexity of big data analysis processing, and there is a limit to preventing/responding to fire in buildings without an integrated platform.

The central government and local governments’ Disaster and Safety Situation Office fails to comprehensively express disaster information between disaster management agencies and conducts disaster management through simple monitoring and risk situation information without smooth communication between disaster management agencies. Decision making by integrated judgment is also delayed due to lack of communication.

In most recent large-scale fires, fire damage is spread due to failure to respond to fire in the initial stage due to failure or stoppage of firefighting facilities in buildings. It is necessary to introduce a management system.

In addition, the current automatic fire alarm system has a limited installation target and is a system that simply transmits only a fire signal. It is necessary to develop a system that automatically provides context information, etc.).

Korean Patent Registration No. 10-1771579 (2017.08.21)

An object of the present invention is to provide a platform that monitors the operation status of firefighting facilities in a building to be monitored in real time by configuring a sound field sensor-based CCTV, a flame image detector, and a detector of a double detection structure for non-fire warning prevention, It is to ensure the reliability of the facility and to prevent and respond to building fires through regular management of whether or not it operates normally.

An object of the present invention is to analyze and predict the state of firefighting facilities received from a fire receiver and feature information extracted from monitoring information (CCTV image) in a building by performing AI-based time-series data analysis and prediction. It is to prevent malfunctions of firefighting facilities in advance in the event of a fire through prediction.

It is an object of the present invention to configure a fire detector according to the installation guide for each type of building and space characteristics, perform situation propagation/broadcast for each fire type in the event of a fire, and transmit 3D scanning space modeling data of a fire-occurring building to a firefighter terminal By doing so, it is possible to suppress the occurrence of non-fire alarms, induce evacuation by fire type, and enable immediate and rapid fire suppression and relief measures by recognizing spatial information in advance by firefighters.

The big data-based building fire prevention response system of the present invention for achieving this technical task is the operating state of firefighting facilities through double detection of receiving status information and sound field information in real time from firefighting facilities provided in buildings to be monitored. a monitoring server that monitors and generates firefighting facility monitoring information; Extract feature data for real fire determination from monitoring information of firefighting facilities, put feature data as input data of neural network, classify fire information and non-fire report through machine learning, and analyze deterioration of firefighting facilities through feature data analysis an analysis server for generating information, replacement time analysis information, and non-fire information analysis information; and a management server that transmits the fire information to the related fire department, indexes the 3D scanning space modeling data of the building in which the fire occurs from the database, and transmits it to the firefighter terminal of the competent fire station.

Preferably, the monitoring server includes: a collection unit for receiving real-time status information and sound field information from firefighting facilities provided in a building to be monitored; And if the status information and sound field information are within the preset reference range, it is determined as a non-fire alarm, and when the status information and the sound field information are out of the preset reference range, the identification ID of the building to be monitored, the identification ID of the firefighting facility, and detection Including a monitoring unit for generating fire-fighting facility monitoring information including time, the state information includes sensor data detected by the fire-fighting facility and flame image data captured through CCTV, and the sound field information is located in a specific space of a building to be monitored. It is characterized in that it includes movement data and temperature change amount data according to the spectral change amount of the generated sound field.

The analysis server includes: an extraction unit for extracting effective data for determining a real fire as feature data from numerical values, texts, images or images included in the firefighting facility monitoring information received from the monitoring server; a learning unit that performs machine learning by deep learning by putting feature data as input data of a neural network, and stores the output data in a database to manage each building or firefighting facility to be monitored; and an analysis unit that compares the characteristic data and the output data to determine that a real fire has occurred in the building to be monitored when the preset deviation is within the reference range and generates fire information.

The management server includes: a notification unit that transmits the fire information received from the analysis server to the associated fire department, and receives the firefighter terminal identification information assigned for fire suppression; a 3D information unit that indexes the 3D scanning space modeling data of the building to be monitored included in the fire information from the database and transmits it to the firefighter terminal; and a facility management unit that classifies the aging and replacement time inspection data received from the analysis server by firefighting facilities, stores and manages them in a database, and updates the replacement cycle for each firefighting facility of the building to be monitored and transmits it to the manager terminal. characterized in that

According to the present invention as described above, by configuring a sound field sensor-based CCTV, a flame image detector, and a detector of a double detection structure for non-fire alarm prevention, a platform for monitoring the operation status of firefighting facilities in a building to be monitored in real time is provided. , it is possible to secure the reliability of firefighting facilities and to prevent and respond to building fires through regular management of normal operation.

According to the present invention, by performing AI-based time-series data analysis and prediction of the state information of the firefighting facility received from the fire receiver and the feature information extracted from the monitoring information (CCTV image) in the building, the deterioration and replacement time of the firefighting facility are predicted. This has the effect of preventing malfunctions of firefighting facilities in advance in the event of a fire.

According to the present invention, the fire detector is configured according to the installation guide for each type of building and space characteristics, but when a fire occurs, situation propagation/broadcasting is performed for each fire type, and 3D scanning space modeling data of the building in which the fire occurred is transmitted to the firefighter terminal. , suppresses the occurrence of non-fire alarms, induces optimal evacuation by fire type, and recognizes firefighters and spatial information to enable immediate and rapid fire suppression and relief measures.

1 is a block diagram illustrating a big data-based building fire prevention response system according to an embodiment of the present invention.
2 is a block diagram illustrating a monitoring server of a big data-based building fire prevention response system according to an embodiment of the present invention.
3 is a block diagram illustrating an analysis server of a big data-based building fire prevention response system according to an embodiment of the present invention.
4 is a block diagram illustrating a fire response management server of a big data-based building fire prevention response system according to an embodiment of the present invention.
5 is a flowchart illustrating a method for preventing fire in a building based on big data according to an embodiment of the present invention.
6 is a flowchart illustrating the detailed process of step S502 of the big data-based building fire prevention response method according to an embodiment of the present invention.
7 is a flowchart illustrating a detailed process of step S504 of the big data-based building fire prevention response method according to an embodiment of the present invention.
8 is a flowchart illustrating the detailed process of step S508 of the big data-based building fire prevention response method according to an embodiment of the present invention.

The specific features and advantages of the present invention will become more apparent from the following detailed description taken in conjunction with the accompanying drawings. Prior to this, the terms or words used in the present specification and claims are based on the principle that the inventor can appropriately define the concept of the term in order to best describe his invention in the technical spirit of the present invention. It should be interpreted with the corresponding meaning and concept. In addition, when it is determined that the detailed description of the well-known functions related to the present invention and its configuration may unnecessarily obscure the gist of the present invention, it should be noted that the detailed description is omitted.

1, the big data-based building fire prevention response system (S) according to an embodiment of the present invention includes a monitoring server 100, an analysis server 200 and a management server 300. is composed

First, the monitoring server 100 receives the state information and the sound field information in real time from the firefighting facilities 11 provided in the building 10 to be monitored through the double detection of the firefighting facilities 11 . It monitors the operation status of firefighting equipment and generates monitoring information for firefighting facilities.

In addition, the analysis server 200 extracts feature data for real fire determination from the firefighting facility monitoring information received from the monitoring server 100, and puts the feature data as input data of the neural network through machine learning to provide fire information and non-fire information and generates aging analysis information, replacement time analysis information, and non-fire information analysis information for the firefighting facility 11 through characteristic data analysis.

And, the management server 300 transmits the fire information received from the analysis server 200 to the associated fire department 20, indexes the 3D scanning space modeling data of the building in which the fire occurred from the database, and configured to transmit to the firefighter terminal 21 .

Hereinafter, a detailed configuration of the big data-based building fire prevention response system (S) according to an embodiment of the present invention will be described with reference to FIGS. 2 to 4 .

Referring to FIG. 2 , the monitoring server 100 includes a collection unit 102 that receives state information and sound field information in real time from firefighting facilities 11 provided in a building 10 to be monitored, and state information and sound field information. If it is within the preset standard range, it is judged to be a non-fire alarm, and if the status information and sound field information are out of the preset standard range, the firefighting facility monitoring including the identification ID of the building to be monitored, the identification ID of the firefighting facility, and the detection time It is configured to include a monitoring unit 104 for generating information.

At this time, the state information includes sensor data detected by the firefighting facility 11 and flame image data captured through CCTV, and the sound field information is movement according to the spectrum change amount of the sound field generated in a specific space of the building 10 to be monitored. data and temperature change data.

In addition, the firefighting facility 11 is equipped with a CCTV, a pressure sensor, a motor voltage sensor, a vibration sensor, a fire hydrant, a hydroxyl device, a transmitter, a temperature sensor, a smoke detector, a gas detector, a sprinkler, a fire door controller, and an alarm provided in the building 10 to be monitored. Alternatively, it is preferable to understand as a generic term for devices that include any one of the fire monitoring control panel and perform a fire monitoring function.

On the other hand, the monitoring unit 104 detects a case in which the data included in the state information authorized from the collection unit 102 abruptly deviates from the preset range while maintaining the set ambient temperature, and instantaneously at a rate of 3 to 8 times per minute. If the power supply to the detector is repeatedly cut off, it is judged as a non-fire alarm. For example, when the relative humidity is rapidly changed 2 to 5 times from 23 ㅁ5% to 90 ㅁ5%, or when the power supply to the sensor is cut off at a rate of 6 times per minute.

In addition, the monitoring unit 104 transmits the inspection information on the firefighting facility 11 matched with the state information that caused the non-fire report to the manager terminal 12, and when both the state information and the sound field information are out of the preset reference range Firefighting facility monitoring information is created in

As such, the monitoring server 100 according to an embodiment of the present invention is constructed with a dual monitoring structure system in consideration of both state information and sound field information, thereby improving the accuracy of fire detection and minimizing malfunction.

The monitoring server 100 matches the term names for each type and function of the firefighting facility 11 in the building 10 to be monitored, and standardizes the data format transmitted by the firefighting facility 11 to the collection unit 102 . It is configured to further include a standardization unit 106 . By configuring the standardization unit 106, it is possible to unify the format of the fire-fighting facility monitoring information for a plurality of buildings 10 to be monitored, thereby facilitating fire monitoring at a remote location.

In addition, the monitoring server 100 according to an embodiment of the present invention may monitor the monitoring target building 10 and the firefighting facility 11 by dividing the monitoring by region, group, or subscriber. For example, in the form of borrowing the service model of security companies such as SECOM, CAPS, it may be configured to perform a series of processes from fire monitoring to response in connection with the competent fire station 20 .

In addition, the monitoring server 100 is developed as an open platform, performs a cloud distributed processing function and a mesh network management function, and supports Android app and tablet monitoring.

In addition, the flame detector, which is one of the firefighting facilities 11, which is the monitoring target of the monitoring server 100, has an independent battery saving function, and enables communication by a short-range network (447 MHz), and has a wavelength of 4.35um to 5.0um. The IR3 sensor that detects and flicker is mounted to detect the flame with a 90 degree sensing angle, and the measurement height is up to 20m and the diameter of the measurement area is configured to have the performance of up to 40m.

In addition, in the detector according to an embodiment of the present invention, a heat detector and a smoke detector are inserted into two slots and connected in parallel, and a unique identification ID is given to identify the location of fire release and double detection function to prevent non-fire alarms and check failures, etc. This is possible.

In addition, the sensor and detector communicate with the MQTT protocol that supports high-speed communication at low capacity suitable for IoT systems and the contact method, and supports the MQTT protocol based on One M2M's Open MTC (Open Machine Type Communications), It can be linked to the smart city standard platform, and sensor data management is possible using the gateway concept, and it is performed by a bridge device connected through HTTP and TCP/IP communication.

Referring to FIG. 3 , the analysis server 200 is an extraction unit that extracts effective data for determining a real fire as characteristic data from numerical values, texts, images or images included in the firefighting facility monitoring information received from the monitoring server 100 . 202, and a learning unit 204 that performs machine learning by deep learning by putting the feature data as input data of the neural network, and stores the output data in a database to manage each building 10 or firefighting facility 11 to be monitored ), and by comparing the characteristic data authorized from the extraction unit 202 with the output data of the learning unit 204, and if the preset deviation is within the reference range, it is determined that a real fire has occurred in the building 10 to be monitored, and fire information is provided. It is configured to include an analysis unit 206 to generate.

At this time, the extraction unit 202 extracts the sensor data out of the set range from the firefighting facility monitoring information, the data digitizing the spectral change of the sound field, and the image or image in which the flame image is taken as characteristic data.

In addition, the learning unit 204 compares the output data for each of the buildings to be monitored 10 for each fire-fighting facility 11, and the identification ID of the fire-fighting facility for the fire-fighting facility 11 indicating a deviation greater than or equal to a preset range, and detection Time is extracted, and old age and replacement time check data for the firefighting facility 11 are generated.

In this way, the learning unit 204 puts the feature data as input data of the neural network and predicts the probability of fire occurrence in Eup / Myeon / Dong unit through the output data derived by performing machine learning by deep learning, It is possible to predict damage, and it is possible to propagate the situation by fire type, operate a building fire/disaster broadcasting system, and operate an intelligent tunnel fire propagation system in connection with real-time public data through Open API.

In addition, the fire information generated by the analysis unit 206 includes detailed information about the building 10 to be monitored (building name, address, location, door location, number of floors, number of households, etc.), details of the firefighting facility 11 installed in the building Any one of information (sprinkler, fire extinguisher, fire hydrant, type of detector, installed/equipped location and number, etc.) or ignition time may be included.

And, the analysis server 200 analyzes the domestic and foreign non-fire information test investigation details and the conventional non-fire information analysis test factor values collected through big data, and conducts a non-fire alarm simulation test according to a preset scenario to determine a reference value for fire. An alarm method in which the shortest evacuation time was derived by analyzing the simulation test unit 208 to update , and further includes a simulation unit 210 that recommends a sensor installation location in consideration of the spatial characteristics of a building according to a simulated fire scenario.

Referring to FIG. 4 , the management server 300 transmits the fire information received from the analysis server 200 to the associated competent fire station 20 , and a notification unit for receiving the firefighter terminal identification information assigned for fire suppression. 302 and the 3D scanning space modeling data of the building 10 to be monitored included in the fire information is indexed from the database and received from the 3D information unit 304 for transmitting to the firefighter terminal 21, and the analysis server 200 Classification of old age and replacement time inspection data for each firefighting facility 11 is stored and managed in the database, and the replacement cycle for each firefighting facility 11 of the building 10 to be monitored is updated to the manager terminal 12. It is configured to include a facility management unit 306 to transmit.

Specifically, the 3D information unit 304 stores and manages 3D spatial data generated through 3D scanning of design drawings or pre-photographed inside the building 10 to be monitored, and VR road view and fire of the building to be monitored. It stores and manages CCTV and VR field scanning data included in the information for each building 10 to be monitored.

In addition, the 3D information unit 304 is a firefighter terminal corresponding to the firefighter terminal identification information authorized by the notification unit 302 at the same time as receiving the fire information. CCTV, and VR are configured to transmit on-site scanning data.

The 3D information unit 304 collects and arranges 3D spatial scanning data (total, inter-floor) of buildings already photographed through the FARO Laser Scanner for each building 10 and firefighting facility 11 to be monitored, and performs data pre-processing in the inter-floor space of the building. The data, the spatial data inside the building, and the spatial data of the building sensor are collected and generated as 3D spatial data.

Accordingly, the 3D information unit 304 detects the fire information and at the same time transmits 3D spatial data matched with the corresponding monitoring target building 10 and the firefighting facility 11 to the firefighter terminal, and accordingly the firefighter generates a fire Fire suppression and relief measures can be implemented while recognizing spatial information.

And, the facility management unit 306 classifies the old age and replacement time inspection data received from the analysis server 200 for each firefighting facility 11, but divides the old age and replacement time inspection data classified for each set monthly inspection cycle to the manager terminal. (12) may be configured to transmit.

In addition, the manager terminal 12 according to an embodiment of the present invention may be composed of any one of a smartphone, a PC, or a PDA that can be connected to the monitoring server 100 , the analysis server 200 or the management server 300 through the information communication network. can

Hereinafter, with reference to FIG. 5, a big data-based building fire prevention response method according to an embodiment of the present invention will be described as follows.

First, the monitoring server 100 collects the state information and sound field information received from the firefighting facilities 11 provided in the building 10 to be monitored to generate firefighting facility monitoring information (S502).

Then, the analysis server 200 analyzes the fire-fighting facility monitoring information and classifies it as fire information or non-fire information (S504).

Subsequently, the management server 300 transmits the fire information to the associated fire department 20 (S506).

Then, the management server 300 indexes the 3D scanning space modeling data of the building in which the fire occurred from the database and transmits it to the firefighter terminal 21 of the competent fire station 20 (S508).

Hereinafter, the detailed process of step S502 of the big data-based building fire prevention response method according to an embodiment of the present invention will be described with reference to FIG. 6 .

First, the monitoring server 100 receives state information and sound field information in real time from the firefighting facilities 11 provided in the building 10 to be monitored (S602).

Subsequently, the monitoring server 100 determines whether the state information and the sound field information are within a preset reference range (S604).

As a result of the determination of step S604, when the state information and sound field information are out of the preset reference range, the monitoring server 100 fires the monitoring information of the fire facilities including the identification ID of the building to be monitored, the identification ID of the fire fighting facility, and the detection time to generate (S606).

Hereinafter, the detailed process of step S504 of the big data-based building fire prevention response method according to an embodiment of the present invention will be described with reference to FIG. 7 .

After step S502, the analysis server 200 receives the firefighting facility monitoring information from the monitoring server 100 (S702).

Next, the analysis server 200 extracts effective data for determining a real fire from numerical values, texts, images or images included in the firefighting facility monitoring information as feature data (S704).

Subsequently, the analysis server 200 performs machine learning by deep learning by putting the feature data as input data of the neural network (S706).

Next, the analysis server 200 compares the feature data and the machine learning output data to determine whether a preset deviation is within a reference range (S708).

As a result of the determination of step S708, if the preset deviation is within the reference range, the analysis server 200 determines that a real fire has occurred in the building 10 to be monitored and generates fire information (S710).

At this time, if the preset deviation is out of the reference range, the analysis server 200 determines that the firefighting facility monitoring information is a non-fire report.

Hereinafter, the detailed process of step S508 of the big data-based building fire prevention response method according to an embodiment of the present invention will be described with reference to FIG. 8 .

After step S506, the management server 300 indexes the 3D scanning space modeling data of the building 10 to be monitored corresponding to the fire information from the database (S802).

Then, the management server 300 receives the fire brigade terminal identification information assigned for fire suppression from the analysis server 200 (S804).

Then, the management server 300 transmits the 3D scanning space modeling data to the firefighter terminal (S806).

As described above, according to the present invention as described above, by configuring a sound field sensor-based CCTV, a flame image detector, and a detector of a double detection structure for preventing non-fire warning, a platform for monitoring the operation status of firefighting facilities in a building to be monitored in real time is provided. By providing it, it is possible to secure the reliability of firefighting facilities and to prevent and respond to building fires through regular management of normal operation.

In addition, by performing AI-based time-series data analysis and prediction on the state information of the firefighting facility received from the fire receiver and the feature information extracted from the monitoring information (CCTV image) in the building, In the event of a fire, it is possible to prevent malfunction of firefighting facilities in advance, and configure the fire detector according to the installation guide for each type of building and space characteristics. By transmitting 3D scanning spatial modeling data, it has the advantage of suppressing the occurrence of non-fire alarms, inducing optimal evacuation by fire type, and enabling immediate and rapid fire suppression and relief measures by recognizing firefighters and spatial information.

Although described and illustrated in relation to a preferred embodiment for illustrating the technical idea of the present invention above, the present invention is not limited to the configuration and operation as shown and described as such, and deviates from the scope of the technical idea. It will be apparent to those skilled in the art that many changes and modifications can be made to the invention without reference to the invention. Accordingly, all such suitable alterations and modifications and equivalents are to be considered as falling within the scope of the present invention.

S: Big data-based building fire prevention response system
10: Building to be monitored
11: Firefighting facilities
12: administrator terminal
20: Fire Department
21: firefighter terminal
100: monitoring server
102: collection unit
104: monitoring unit
106: standardization unit
200: analysis server
202: extraction unit
204: study department
206: analysis unit
300: management server
302: notification unit
304: 3D intelligence department
306: facility management department

Claims (4)

a monitoring server 100 for generating monitoring information for firefighting facilities by monitoring the operation status of firefighting facilities through dual sensing for receiving state information and sound field information in real time from firefighting facilities provided in a building to be monitored;
Extracting feature data for real fire determination from the firefighting facility monitoring information, putting the feature data as input data of a neural network, classifying it into fire information and non-fire information through machine learning, and analyzing the characteristics of the firefighting facilities Analysis server 200 for generating analysis information, replacement time analysis information, and non-fire information analysis information; and
A management server that transmits the fire information to the related fire department 20, indexes the 3D scanning space modeling data of the building in which the fire occurs from the database and transmits it to the firefighter terminal of the competent fire station,
The monitoring server 100 matches the term names for each type and function of the firefighting facility 11 in the building 10 to be monitored, and standardizes the data format transmitted by the firefighting facility 11 to the collection unit 102 . part (106);
The analysis server 200 analyzes the domestic and foreign non-fire warning test investigation details and the conventional non-fire warning analysis test factor values collected through big data, and conducts a non-fire simulation test according to a preset scenario to determine a reference value for fire determination. The simulation test unit 208 to update, and the evacuation time pattern required according to the priority and all-time alarm methods of the fire alarm system are analyzed by building type and fire location, and the shortest evacuation time is derived as an alarm method. Big data-based building fire prevention response system, characterized in that it includes a simulation unit 210 that sets and recommends a detector installation location in consideration of the spatial characteristics of a building according to a simulated fire scenario. According to claim 1,
The monitoring server,
a collection unit for receiving real-time status information and sound field information from firefighting facilities provided in the building to be monitored; and
If the status information and sound field information are within the preset reference range, it is determined that there is a non-fire alarm. Including a monitoring unit for generating fire-fighting facility monitoring information including time,
The state information includes sensor data detected by the firefighting facility and flame image data captured through CCTV, and the sound field information includes movement data and temperature change data according to the spectrum change amount of the sound field generated in a specific space of the building to be monitored. Big data-based building fire prevention response system, characterized in that it includes. According to claim 1,
The analysis server,
an extraction unit for extracting effective data for real fire determination from numerical values, texts, images or images included in the firefighting facility monitoring information received from the monitoring server as feature data;
a learning unit that performs machine learning by deep learning by putting the feature data as input data of a neural network, and stores the output data in a database to manage each building or firefighting facility to be monitored; and
An analysis unit that compares the characteristic data with the output data and generates fire information by determining that a real fire has occurred in the building to be monitored when the preset deviation is within the reference range
Big data-based building fire prevention response system, characterized in that it includes. According to claim 1,
The management server,
a notification unit for transmitting the fire information received from the analysis server to a related fire department, and receiving identification information of a firefighter terminal assigned for fire suppression;
a 3D information unit that indexes 3D scanning space modeling data of a building to be monitored included in the fire information from a database and transmits it to a firefighter terminal; and
A facility management unit that classifies the aging and replacement time inspection data received from the analysis server by firefighting facilities, stores and manages them in a database, and updates the replacement cycle for each firefighting facility of the building to be monitored and transmits it to the manager terminal
Big data-based building fire prevention response system, characterized in that it includes.

Images