KR20230126026A - Control server, method and computer program to detect fire - Google Patents

Abstract

A fire detection method performed by a control server for detecting a fire in a target area includes collecting sensor data for at least one sensor from a collection server, and generating the sensor data based on characteristics of the sensor data when the sensor data is collected. Clustering into a plurality of clusters, selecting a specific cluster from among the plurality of clusters based on current status information of the target area, receiving real-time sensor data for the at least one sensor from the collection server, and and determining whether a fire has occurred in the target area based on whether real-time sensor data exceeds a predetermined threshold and a similarity between the real-time sensor data and the specific cluster.

Description

Control server, method and computer program for detecting fire {CONTROL SERVER, METHOD AND COMPUTER PROGRAM TO DETECT FIRE}

The present invention relates to a control server, method and computer program for detecting a fire.

Conventional fire detection and alarm systems are mostly sensor-based detectors such as heat and smoke, and such sensor-based detectors detect the occurrence of fire only when heat or smoke spreads to a certain level or reaches or flows into the sensor after a fire occurs. can do.

A threshold value (upper limit) was set for the data collected from the fire detector, and a fire was judged by logic such as when the values of each heat, smoke, and flame data exceed one or more than two threshold values. However, this method of determining fire does not reflect the characteristics of complex data depending on the existence of people in the place where it is installed or the characteristics of the place, which is the main cause of generating non-fire alarms.

In addition, when a person is always present, even if a fire occurs, it can be immediately recognized, and since radio waves are possible through mobile phones and voice calls, filtering according to the purpose of the installation place or operating time is required. For example, fire detectors are often turned off due to frequent non-fire alarms (misinformation) in spaces with a lot of smoke or high temperature due to the nature of facilities such as cooking, bathing facilities, small-scale manufacturing, factories, and logistics facilities. In case of an outbreak, the initial response is inevitably insufficient, which can lead to a large-scale fire accident.

The present invention is to provide a control server, method and computer program for detecting a fire.

However, the technical problem to be achieved by the present embodiment is not limited to the technical problems described above, and other technical problems may exist.

As a means for achieving the above-described technical problem, an embodiment of the present invention includes the steps of collecting sensor data for at least one sensor from a collection server, based on the sensor data characteristics at the time of collecting the sensor data Clustering sensor data into a plurality of clusters, selecting a specific cluster from among the plurality of clusters based on current status information of the target area, and receiving real-time sensor data for the at least one sensor from the collection server. and determining whether a fire has occurred in the target area based on whether the real-time sensor data exceeds a preset threshold and a similarity between the real-time sensor data and the specific cluster. can

Another embodiment of the present invention is a collection unit that collects sensor data for at least one sensor from a collection server, and clustering that clusters the sensor data into a plurality of clusters based on characteristics of sensor data when the sensor data is collected. a cluster selector that selects a specific cluster from among the plurality of clusters based on current status information of the target area; a receiver that receives real-time sensor data for the at least one sensor from the collection server; and the real-time sensor data The control server may include a fire determination unit configured to determine whether or not a fire occurs in the target area based on whether a preset threshold is exceeded and a similarity between the real-time sensor data and the specific cluster.

In another embodiment of the present invention, a computer program, when executed by a computing device, collects sensor data for at least one sensor from a collection server, and collects the sensor data based on characteristics of the sensor data when the sensor data is collected. Clustering data into a plurality of clusters, selecting a specific cluster from among the plurality of clusters based on current status information of the target area, receiving real-time sensor data for the at least one sensor from the collection server, and receiving the real-time sensor data from the real-time sensor A computer program stored in a computer readable recording medium including a sequence of instructions for determining whether a fire occurs in the target area based on whether data exceeds a predetermined threshold and the similarity between the real-time sensor data and the specific cluster can provide.

The above-described means for solving the problems is only illustrative and should not be construed as limiting the present invention. In addition to the exemplary embodiments described above, there may be additional embodiments described in the drawings and detailed description.

According to any one of the above-described problem solving means of the present invention, by clustering sensor data of a target area into a plurality of clusters and detecting a fire using a specific cluster based on current status information of the target area and real-time sensor data, Since the characteristics of the area can be reflected, fire detection accuracy can be increased and false alarms can be reduced.

1 is a schematic diagram of a fire detection system according to an embodiment of the present invention.
2 is a configuration diagram of a control server according to an embodiment of the present invention.
3 is an exemplary diagram of sensor data collected from a collection server according to an embodiment of the present invention.
4A and 4B are exemplary diagrams for explaining a plurality of clustered clusters according to an embodiment of the present invention.
5 is an exemplary diagram for explaining a fire detection method using a clustered model according to an embodiment of the present invention.
6 is a flowchart of a fire detection method performed in a control server according to an embodiment of the present invention.

Hereinafter, embodiments of the present invention will be described in detail so that those skilled in the art can easily practice the present invention with reference to the accompanying drawings. However, the present invention may be embodied in many different forms and is not limited to the embodiments described herein. And in order to clearly explain the present invention in the drawings, parts irrelevant to the description are omitted, and similar reference numerals are attached to similar parts throughout the specification.

Throughout the specification, when a part is said to be "connected" to another part, this includes not only the case where it is "directly connected" but also the case where it is "electrically connected" with another element interposed therebetween. . In addition, when a part "includes" a certain component, this means that it may further include other components, not excluding other components, unless otherwise stated, and one or more other characteristics. However, it should be understood that it does not preclude the possibility of existence or addition of numbers, steps, operations, components, parts, or combinations thereof.

In this specification, a "unit" includes a unit realized by hardware, a unit realized by software, and a unit realized using both. Further, one unit may be realized using two or more hardware, and two or more units may be realized by one hardware.

In this specification, some of the operations or functions described as being performed by a terminal or device may be performed instead by a server connected to the terminal or device. Likewise, some of the operations or functions described as being performed by the server may also be performed in a terminal or device connected to the corresponding server.

Hereinafter, an embodiment of the present invention will be described in detail with reference to the accompanying drawings.

1 is a schematic diagram of a fire detection system according to an embodiment of the present invention.

Referring to FIG. 1 , a fire detection system 1 according to an embodiment of the present invention may include at least one sensor 10, a collection server 100, and a control server 200. At least one sensor 10, a collection server 100, and a control server 200 shown in FIG. 1 illustrate components that can be controlled by the fire detection system 1 by way of example.

Each component of the fire detection system 1 of FIG. 1 may be generally connected through a network. For example, a network refers to a connection structure capable of exchanging information between nodes such as terminals and servers, such as a local area network (LAN), a wide area network (WAN), and the Internet. (WWW: World Wide Web), including wired and wireless data communications networks, telephone networks, and wired and wireless television communications networks. Examples of wireless data communication networks include 3G, 4G, 5G, 3rd Generation Partnership Project (3GPP), Long Term Evolution (LTE), World Interoperability for Microwave Access (WIMAX), Wi-Fi, Bluetooth communication, infrared communication, ultrasonic communication, visible light communication (VLC: Visible Light Communication), LiFi, and the like, but are not limited thereto.

The collection server 100 may collect sensor data for at least one sensor 10 . The sensor data may include data collected from each of the at least one sensor 10 .

Here, the at least one sensor 10 may include a sensor capable of detecting at least one of smoke, temperature, and flame. However, it is not limited thereto, and the at least one sensor 10 may include various sensors capable of collecting fire-related data.

At least one sensor 10 may be an analog (address type) sensor or an IoT type sensor. At this time, the IoT-type sensor is connected to a gateway wirelessly or wired, and sensor data may be transmitted to the collection server 100 through such a gateway. Unlike sensors using existing electric wires, the at least one sensor 10 may periodically transmit sensor data to the collection server 100 to determine whether or not there is a fire.

The control server 200 may detect a fire in the target area using sensor data received from the collection server 100 . Here, the target area is a target area for determining whether a fire is detected, and may include, for example, a specific building, a specific floor, and a specific space (eg, a conference room, a bathroom, etc.). At least one sensor 10 may be installed in the target area to detect fire.

The control server 200 may collect sensor data for at least one sensor 10 from the collection server 100 .

The control server 200 may cluster sensor data into a plurality of clusters based on sensor data characteristics when sensor data is collected.

The control server 200 may select a specific cluster from among a plurality of clusters based on current status information of the target area.

The control server 200 may receive real-time sensor data for at least one sensor 10 from the collection server 100 .

The control server 200 may determine whether a fire occurs in the target area based on whether the real-time sensor data exceeds a preset threshold and the similarity between the real-time sensor data and a specific cluster.

When the control server 200 determines that a fire has occurred, it can be transmitted to the control center 20 so that an alarm screen is displayed along with an alarm sound. The operator of the control center 20 can check this and make a wired call to the related terminal 30 or check the CCTV installed in the target area to determine the occurrence of a fire.

In addition, the control server 200 may send a fire notification to the manager terminal 30 through the control center 20 using multimedia such as SMS, PUSH message using APP, or WEB.

In addition, the control server 200 may send a report message to the fire department server 40 when a predetermined reporting condition is satisfied, and report the fire department headquarters situation room 50 to be dispatched to the target area. At this time, the report message may include information necessary for initial dispatch.

In addition, the control server 200 may control the gas fire extinguishing facility 60 and the fire shutter 70 so that the gas fire extinguishing facility 60 and the fire shutter 70 operate automatically when a fire is determined. The gas fire extinguishing facility 60 may extinguish the fire by lowering the oxygen concentration in the area where the fire occurs under the control of the control server 200 . The fire shutter 70 may be partially or completely lowered under the control of the control server 200 .

According to the fire detection system 1 according to an embodiment of the present invention, sensor data of a target area when a fire does not occur is clustered into a plurality of clusters, and specific clusters and real-time sensor data based on current status information of the target area are clustered. By detecting a fire using , it is possible to reflect the characteristics of the target area, thereby increasing the accuracy of fire detection and reducing false alarms.

That is, by reducing the non-fire alarm by reflecting the characteristics of the target area, it is possible to maintain the normal operation of the fire protection system (fire shutter or gas fire extinguishing equipment, etc.) in the event of a fire by reducing cases in which the person concerned artificially turns off the fire detection system. In addition, it is possible to prevent secondary damage caused by the fire-fighting system by preventing accidents of the fire shutter due to the false alarm of the fire detector and suffocation due to the operation of the carbon dioxide fire extinguishing equipment. In addition, by transmitting a report message including information necessary for initial dispatch to the fire station when it is determined that a fire has occurred, the fire station can facilitate the initial response to the fire.

2 is a configuration diagram of a control server according to an embodiment of the present invention.

Referring to FIG. 2 , the control server 200 according to an embodiment of the present invention includes a collection unit 210, a clustering unit 220, a cluster selection unit 230, a receiving unit 240, and a fire determination unit 250. And it may include a fire reporting unit (260). However, the control server 200 shown in FIG. 2 is only one implementation example of the present invention, and various modifications are possible based on the components shown in FIG. 2 .

Hereinafter, FIG. 1 will be described with reference to FIGS. 2 to 4 .

The collection unit 210 may collect sensor data for at least one sensor 10 from the collection server 100 . The sensor data may include data collected from each of the at least one sensor 10 . For example, the sensor data may include at least one of a unique identifier of the corresponding sensor, a measurement value of the corresponding sensor, and a measurement time.

In one embodiment, the at least one sensor 10 may include a sensor installed in the target area and capable of detecting at least one of smoke, temperature, and flame, respectively. For example, when at least one sensor 10 includes a smoke sensor, a temperature sensor, and a flame sensor, the sensor data includes a unique identifier of each sensor, a value measured by detecting smoke, temperature, and flame from each sensor, and a measured value. may include time. Here, the target area may be a target area for determining whether a fire is detected.

At this time, the collected sensor data may be decomposed into time-series data streams in accordance with standards. For example, when the sensor 10 is composed of a composite sensor capable of detecting smoke, temperature, and flame, as shown in FIG. 3 , the sensor data collected from the collection server 100 is a unique identifier of the corresponding sensor. 310 , a smoke measurement value 320 , a temperature measurement value 330 , a flame measurement value 340 , and a time 350 when the corresponding measurement value was detected.

Here, when the collection server 100 collects sensor data from at least one sensor 10 , the collected sensor data may be decomposed into time-series data streams in accordance with standards and transmitted to the collection unit 210 . However, it is not limited thereto, and the collection unit 210 may directly decompose the sensor data collected from the collection server 100 into a time-series data stream form in accordance with the standard.

The clustering unit 220 may cluster sensor data into a plurality of clusters based on characteristics of sensor data when sensor data is collected.

Here, clustering (cluster analysis) is a machine learning technique that groups data with similar characteristics into the same group, and at this time, a cluster may be a group of data with similar characteristics.

In this regard, unsupervised learning may be used for clustering because it is common to group unlabeled data.

In one embodiment, the clustering unit 220 may cluster sensor data into a plurality of clusters using a density-based clustering method. The density-based clustering method is based on the assumption that 'data belonging to the same cluster will be distributed close to each other', and can be a method of recognizing as one cluster if there are n or more points within a certain radius based on a point. . In the case of center-based algorithms such as k-means or hierarchical clustering, clustering is performed using the distance between clusters, whereas in case of density-based clustering, points are closely clustered, It is a method of clustering high-density parts.

The clustering unit 220 may perform clustering using a Density-based spatial clustering of applications with noise (DBSCAN) algorithm among density-based clustering methods. Regarding this DBSCAN algorithm, a function representing the number of points (Number of points) included in a cluster corresponding to a radius of Eps based on a random point q included in the database D is defined as in Equation 1 below. can

where D is a database, p is any point, q is a neighboring point of point p, Eps is the maximum radial distance that can include neighboring points in a cluster, and dist(p, q) can be the distance between points p and q. .

In an embodiment, the sensor data characteristics may include at least one of time zone, season, temperature, and day of the week when sensor data is collected. In addition, the sensor data characteristic may include a characteristic that may reflect at least one of a characteristic of a place where a sensor is installed, a user's work pattern, and a user's space usage pattern.

Specifically, the clustering unit 220 may cluster the sensor data into a plurality of clusters by reflecting the characteristics of the sensor data when there is no fire. For example, the clustering unit 220 may cluster the sensor data into two clusters by reflecting a case where a person exists and a case where a person does not exist in the target area, and reflects time zones such as morning, afternoon, and night. Accordingly, the sensor data may be clustered into three clusters, and the clustering may be further classified and clustered by reflecting the characteristics of two or more sensor data.

In an embodiment, the clustering unit 220 may cluster sensor data into a plurality of clusters according to time zones based on a distribution map of people according to time zones. For example, as shown in FIGS. 4A and 4B , the clustering unit 220 may classify and cluster sensor data of night time zones when no people are present and morning and afternoon time zones when people are present. Specifically, the plurality of clusters clustered based on sensor data of flame, smoke, and temperature include a first cluster 410 and 440 including each sensor data of the morning time zone and a first cluster 410 and 440 including each sensor data of the afternoon time zone. It may include the second clusters 420 and 450 and the third clusters 430 and 460 including each sensor data of the night time zone.

In one embodiment, the clustering unit 220 may cluster sensor data into a plurality of clusters based on characteristics of a place where the sensor is installed. For example, the clustering unit 220 determines the operation time of the facility or the facility operation time of the facility for a place with a lot of smoke or high temperature due to the nature of the facility, such as cooking, bathing facilities, small-scale manufacturing facilities, factories, and logistics facilities. Sensor data may be clustered into a plurality of clusters.

In one embodiment, the clustering unit 220 may cluster the sensor data into a plurality of clusters based on the user's work pattern or the user's space use pattern. The clustering selection unit 230 may determine whether a corresponding user exists in a target area by using the location of a specific user's mobile terminal and whether the mobile terminal is connected to a router, and analyze the working pattern or space usage pattern of the corresponding user. .

For example, the clustering unit 220 may cluster the sensor data into a plurality of clusters according to whether a specific user performs a job generating flame, smoke, or temperature and a time pattern for performing the corresponding job. Alternatively, the clustering unit 220 may cluster sensor data into a plurality of clusters according to a pattern used by a specific user only in a specific target region.

In this way, by clustering into a plurality of clusters according to the characteristics of the sensor data, non-fire alarm is provided even in a place where there is a lot of smoke or high temperature due to the nature of the place, or even in a situation where flames, smoke, or temperature may occur intermittently even when there is no fire. can prevent this from happening.

The clustering unit 220 may periodically collect sensor data and corresponding sensor data characteristics to update a clustered model.

The cluster selector 230 may select a specific cluster from among a plurality of clusters based on current status information of the target region.

In this case, the current status information of the target area may include at least one of time zone, season, temperature and day of the week, a place where at least one sensor is installed, a user's work pattern, and a user's space usage pattern. For example, when sensor data for at least one sensor installed in the target area is clustered into three clusters corresponding to night, morning, and afternoon time zones, the cluster selector 230 determines a fire among the three clusters. You can select the cluster of the corresponding time zone.

As another example, if sensor data for at least one sensor installed in the target area is clustered into two clusters: a case where a person exists in the target area and a case where a person does not exist, whether a person exists in the target area Depending on this, you can select a specific cluster.

At this time, the cluster selector 230 may determine whether a user exists in the target area based on the location of at least one user's mobile terminal related to the target area and whether each mobile terminal is connected to a router installed in the target area. and, accordingly, a specific cluster can be automatically selected.

For example, the cluster selection unit 230 uses the current time and the location of the user's mobile terminal to determine if the mobile terminal does not exist in the target area for a certain period of time or if each mobile terminal is not connected to a router installed in the target area. It may be determined that the user is not located in the target area. In this case, the cluster selector 230 may select a cluster corresponding to a case where no person exists in the target area.

In one embodiment, the clustering unit 220 may cluster sensor data into a plurality of clusters according to the distribution of people. Accordingly, the cluster selection unit 230 determines how many people are distributed in the target area by using the location of at least one user's mobile terminal related to the target area and whether each mobile terminal is connected to a router installed in the target area. can The cluster selector 230 may select a specific cluster from among a plurality of clusters clustered according to the distribution of people in the target area.

The receiver 240 may receive real-time sensor data for at least one sensor 10 from the collection server 100 . The real-time sensor data may include at least one of a unique identifier of the corresponding sensor, a measurement value of the corresponding sensor, and a measurement time.

The fire determination unit 250 may determine whether a fire has occurred in the target area based on whether the real-time sensor data exceeds a preset threshold and the similarity between each real-time sensor data and a specific cluster.

Specifically, the fire determination unit 250 may determine that a fire has occurred in the target area when the real-time sensor data exceeds a predetermined threshold and the real-time sensor data does not fall within a range of a specific cluster.

For example, the fire determination unit 250 may determine that no fire has occurred in the corresponding target region if the real-time sensor data is within the range of a specific cluster in the same time zone of the previously learned target region.

For example, if the threshold for temperature is X, the threshold for smoke is Y, and the threshold for flame is Z, a set of real-time sensor data transmitted from each sensor The fire determination method of the fire determination unit 250 for is as follows CASE 1.

According to CASE 1, if at least one of real-time sensor data for temperature, smoke, and flame exceeds a predetermined threshold and the real-time sensor data does not fall within the range of a specific cluster, the fire determination unit 250 determines the target area. It can be judged that a fire has occurred.

As another example, as in the following CASE 2, the fire determination unit 250 determines whether the real-time sensor data for temperature, smoke, and flame exceed a predetermined threshold and at the same time the real-time sensor data does not fall within the range of a specific cluster. In this case, it can be determined that a fire has occurred in the target area.

In the case of fire shutters or gas fire extinguishing equipment linked to sensor data, there is a concern about human life damage in case of malfunction due to non-fire alarm (false alarm). can

The fire reporting unit 260 may report the occurrence of a fire in a corresponding target area when the magnitude of the vector of the real-time sensor data increases during a preset period or by a preset number of times.

The fire reporting unit 260 measures each of the measured values of temperature, smoke, and flame included in the real-time sensor data. , the magnitude d of the vector of real-time sensor data can be expressed as in Equation 2 below.

For example, the fire reporting unit 260 may report the occurrence of a fire in a corresponding target area when the vector size of the real-time sensor data increases five times or more consecutively or within one minute.

The fire reporting unit 260 uses sensor data information and user information to provide not only the exact location of the fire, but also information necessary for the initial dispatch of the fire department, such as the presence or absence of dangerous materials, the number of people present in the place, and the number of fire floors (height). You can create and report a fire outbreak.

According to one embodiment of the present invention, the role of automatically reporting to the fire department when a fire is detected, which is the role of a fire alarm of an existing fire court facility, can be acted as an agent. Conventional fire alarm systems report the occurrence of a fire by sending recorded voice information or simple fire location information to the fire station after detecting a fire. Frequent non-fire reports are an important cause of wasting firefighting power in case of automatic reporting to the fire department.

Therefore, according to an embodiment of the present invention, it is not reported even if the conditions of CASE 1 or CASE 2 described above are satisfied, and when the condition of Equation 2 is satisfied, it is reported to the fire department to reduce the misfire of the fire department and waste fire power can stop

5 is an exemplary diagram for explaining a fire detection method using a clustered model according to an embodiment of the present invention. At this time, it is assumed that the measured values for smoke, temperature, and flame included in the sensor data are clustered into three clusters based on time zones.

Referring to FIG. 5 , the receiver 240 may receive real-time sensor data for each of the at least one sensor 10 from the collection server 100 (S510). For example, as shown in FIG. 5 , when at least one sensor 10 is composed of a composite sensor capable of detecting smoke, temperature, and flame, the received real-time sensor data includes sensor ID, smoke measurement value, It may include temperature measurements and flame measurements.

Subsequently, the fire determination unit 250 may analyze fire detection based on the received real-time sensor data (S520). Specifically, the fire determination unit 250 may analyze the similarity between the received real-time sensor data and the cluster selected through the cluster selection unit 230 .

That is, the cluster selector 230 may select a specific cluster among three clusters (Cluster 1, Cluster 2, and Cluster 3) based on the current status information of the target region (S530). As shown in FIG. 5 , the three clusters may include cluster 1, cluster 2, and cluster 3. For example, cluster 1 may include sensor data in the night time zone, cluster 2 may include sensor data in the morning time zone, and cluster 3 may include sensor data in the afternoon time zone. If the current time zone is the night time zone, the cluster selector 230 may select cluster 1 corresponding to the night time zone.

Subsequently, the fire determination unit 250 may determine whether a fire occurs in the target area (S540). As described above, the fire determination unit 250 may determine whether a fire occurs in the target area based on whether the real-time sensor data exceeds a predetermined threshold and the similarity between each real-time sensor data and a specific cluster.

When it is determined that no fire has occurred in the target area (S540, No), the fire determination unit 250 may analyze fire detection based on the received real-time sensor data (S520).

When it is determined that a fire has occurred in the target area (S540, Yes), the fire determination unit 250 may send a fire notification to the related terminal 30 or notify the control center 20 that a fire has occurred in the target area. Accordingly, the operator of the control center 20 may confirm the occurrence of a fire by making a wired call to the related terminal 30 .

In addition, when the fire determination unit 250 determines that a fire has occurred in the target area (S540, Yes), the fire reporting unit 260 may analyze real-time sensor data (S550). Specifically, the fire reporting unit 260 may calculate the size of a vector of real-time sensor data.

In addition, the fire reporting unit 260 may determine whether a preset condition for reporting a fire is satisfied (S560). As described above, the fire reporting unit 260 may determine whether the vector size of the real-time sensor data increases during a preset period or by a preset number of times.

If the fire reporting unit 260 does not satisfy the preset conditions (S560, No), it may continuously analyze the real-time sensor data (S550).

The fire reporting unit 260 may report the occurrence of a fire in a corresponding target area by sending a reporting message 51 when a preset condition is satisfied (S560, Yes). At this time, the report message 51 is information on the target area, that is, GPS information, store name, store phone number, store owner information, store location, dangerous material information, suspected fire information (temperature, flame, smoke, etc.), fire occurrence determination It may include at least one of time and control center information of a corresponding target area.

6 is a flowchart of a fire detection method performed in a control server according to an embodiment of the present invention. The fire detection method performed by the control server 200 shown in FIG. 6 includes steps processed time-sequentially according to the embodiment shown in FIGS. 1 to 5 . Therefore, even if the content is omitted below, it is also applied to the fire detection method performed in the control server 200 according to the embodiment shown in FIGS. 1 to 5 .

In step S610, the control server 200 may collect sensor data for at least one sensor from the collection server.

In step S620, the control server 200 may cluster each sensor data into a plurality of clusters based on sensor data characteristics when the sensor data is collected.

In step S630, the control server 200 may select a specific cluster from among a plurality of clusters based on current status information of the target area.

In step S640, the control server 200 may receive real-time sensor data for at least one sensor from the collection server.

In step S650, the control server 200 may determine whether a fire has occurred in the target area based on whether the real-time sensor data exceeds a preset threshold and the similarity between each real-time sensor data and a specific cluster.

In the above description, steps S610 to S650 may be further divided into additional steps or combined into fewer steps, depending on an embodiment of the present invention. Also, some steps may be omitted as needed, and the order of steps may be switched.

The fire detection method performed by the control server 200 described with reference to FIGS. 1 to 6 may be implemented in the form of a computer program stored in a medium executed by a computer or a recording medium including instructions executable by a computer. there is. In addition, the fire detection method performed by the control server 200 described with reference to FIGS. 1 to 6 may be implemented in the form of a computer program stored in a medium executed by a computer.

Computer readable media can be any available media that can be accessed by a computer and includes both volatile and nonvolatile media, removable and non-removable media. Also, computer readable media may include computer storage media. Computer storage media includes both volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data.

The above description of the present invention is for illustrative purposes, and those skilled in the art can understand that it can be easily modified into other specific forms without changing the technical spirit or essential features of the present invention. will be. Therefore, the embodiments described above should be understood as illustrative in all respects and not limiting. For example, each component described as a single type may be implemented in a distributed manner, and similarly, components described as distributed may be implemented in a combined form.

The scope of the present invention is indicated by the following claims rather than the detailed description above, and all changes or modifications derived from the meaning and scope of the claims and equivalent concepts should be interpreted as being included in the scope of the present invention. do.

10: sensor
100: collection server
200: control server
210: collection unit
220: clustering unit
230: cluster selection unit
240: receiver
250: fire judgment unit
260: fire reporting department

Claims (13)

A fire detection method performed in a control server for detecting a fire in a target area,
collecting sensor data for at least one sensor from a collection server;
clustering the sensor data into a plurality of clusters based on characteristics of the sensor data when the sensor data is collected;
selecting a specific cluster from among the plurality of clusters based on current status information of the target region;
receiving real-time sensor data for the at least one sensor from the collection server; and
Determining whether a fire has occurred in the target area based on whether the real-time sensor data exceeds a preset threshold and a similarity between the real-time sensor data and the specific cluster
To include, a fire detection method.
According to claim 1,
The sensor data characteristic includes at least one of the time zone, season, temperature and day of the week when the sensor data is collected, a place where the at least one sensor is installed, a user's work pattern, and a user's space use pattern. detection method.
According to claim 1,
Clustering the sensor data into a plurality of clusters,
Clustering the sensor data into a plurality of clusters using a density-based clustering method.

According to claim 1,
The step of selecting the specific cluster,
and automatically selecting the specific cluster based on the location of at least one user's mobile terminal related to the target area and whether each mobile terminal is connected to a router installed in the target area. method.
According to claim 1,
The step of determining whether a fire occurs in the target area,
and determining that a fire has occurred in the target area when the real-time sensor data exceeds a preset threshold and the real-time sensor data falls within a range of the specific cluster.
According to claim 1,
Reporting the occurrence of a fire in the target area when the size of the vector of the real-time sensor data increases during a predetermined period or by a predetermined number of times
To further include, a fire detection method.
In the control server for detecting a fire in a target area,
a collection unit that collects sensor data for at least one sensor from the collection server;
a clustering unit clustering the sensor data into a plurality of clusters based on characteristics of the sensor data when the sensor data is collected;
a cluster selector selecting a specific cluster from among the plurality of clusters based on the current state information of the target region;
a receiving unit receiving real-time sensor data for the at least one sensor from the collection server; and
A fire determination unit for determining whether a fire has occurred in the target area based on whether the real-time sensor data exceeds a preset threshold and a similarity between the real-time sensor data and the specific cluster
To include, the control server.
According to claim 7,
The sensor data characteristic includes at least one of the time zone, season, temperature and day of the week when the sensor data is collected, a place where the at least one sensor is installed, a user's work pattern, and a user's space use pattern. server.
According to claim 7,
Wherein the clustering unit clusters the sensor data into a plurality of clusters using a density-based clustering method.
According to claim 7,
wherein the cluster selection unit automatically selects the specific cluster based on the location of at least one user's mobile terminal related to the target area and whether each mobile terminal is connected to a router installed in the target area. .
According to claim 7,
Wherein the fire determination unit determines that a fire has occurred in the target area when the real-time sensor data exceeds a preset threshold and the real-time sensor data falls within a range of the specific cluster.
According to claim 7,
A fire reporting unit for reporting the occurrence of a fire in the target area when the size of the vector of the real-time sensor data increases during a predetermined period or by a predetermined number of times
To further include, the control server.
A computer program stored in a computer readable recording medium including a sequence of instructions for detecting a fire in a target area,
When the computer program is executed by a computing device,
Collecting sensor data for at least one sensor from the collection server;
Clustering the sensor data into a plurality of clusters based on characteristics of the sensor data when each sensor data is collected;
Selecting a specific cluster from among the plurality of clusters based on the current status information of the target region;
receiving real-time sensor data for the at least one sensor from the collection server;
A computer readable recording medium comprising a sequence of instructions for determining whether a fire has occurred in the target area based on whether the real-time sensor data exceeds a predetermined threshold and a similarity between the real-time sensor data and the specific cluster stored computer programs.

Images