KR101903582B1 - Led lighting system with disaster prevention detection and disaster prevention training function, and a method thereof - Google Patents

Abstract

The present invention relates to an LED lighting system with a disaster detection and disaster prevention training function and a method thereof. More particularly, the LED lighting system includes a disaster detection and control device and an LED lighting unit including a plurality of sensors in each office (store) in a building or a predetermined area. It is possible to prevent the occurrence of disaster by analyzing the sensed data collected through the plurality of sensors using a data mining algorithm and a situation recognition analysis algorithm. It is possible to prevent damage caused by fire by generally blocking power when a disaster occurs. It is possible to conduct disaster evacuation drills and minimize casualties at the disaster by quickly evacuating people through the output color, the output light or the output pattern, or the sound output of an LED lighting unit according to a disaster training control signal, a disaster prevention control signal or a disaster occurrence control signal.

Description

TECHNICAL FIELD [0001] The present invention relates to an LED lighting system having a disaster prevention detection function and a disaster training function,

The present invention relates to an LED lighting system and a method thereof having a disaster prevention detection function and a disaster training function, and more particularly, to an LED lighting system having a disaster detection system including a plurality of sensors in each office (store) And a controller and an LED illumination unit. By analyzing the sensed data collected through the plurality of sensors by using a data mining algorithm and a situation recognition analysis algorithm, it is possible to prevent the occurrence of a disaster in advance, Thus, it is possible to minimize the damage caused by the fire and quickly evacuate through the output color, the output light amount or the output pattern of the LED lighting unit or the sound output according to the disaster training control signal, disaster prevention control signal or disaster occurrence control signal By doing so, it is possible to disaster evacuation drills and minimize disaster And an LED lighting system and a method thereof, which are equipped with an emergency occurrence detection function and a disaster training function.

In recent years, as urban densification has become more common, there are increasing trends in multi-purpose high-rise buildings where commercial housing, accommodation, and sales facilities are mixed in business facilities. In addition to high-rise apartment buildings in residential areas, high-rise residential and commercial complex buildings are also increasing in commercial areas, and multi-use buildings such as cultural gathering facilities, accommodation facilities, amusement facilities including sales facilities It is increasing rapidly.

On the other hand, accident rates due to disasters increase by 5% each year, but the prevention and safeguards against such disasters still use the past methods.

In other words, the development of electronic and measurement technology has led to the use of automated measurement management systems that use state-of-the-art devices and methods in many fields. However, in addition to natural disasters such as typhoons and floods, The field is still maintaining the system that the manager supervises directly.

In this way, direct management supervision by the manager is not always in itself, and because the manpower is not supported much because the manpower is not supported, a system that can monitor the disaster 24 hours without dependency on the manpower is required.

In addition, the existing facilities for disaster management in buildings are all the fire-fighting facilities specified in the Fire Service Act, and they are trained simply by facility managers and workers' manuals and responded to the subjective judgment of managers and workers in case of an accident or disaster to be. There are no dedicated systems and software for disaster management.

Therefore, it may be difficult to predict the disaster management and to respond in the early stage due to the absence of facility detection system for disaster management to predict and respond to disasters.

On the other hand, disaster disasters such as large-scale fire, gas leakage, flooding, collapse, and traffic accident have occurred more frequently in recent years. To cope with this, it is necessary to continually develop equipment that detects the occurrence of a disaster in a specific space, such as a building, a subterranean space, or a moving transportation means, and alerts the surroundings.

Until recently, it was the most active fire prevention measure in Korea to install a short circuit breaker to prevent fire caused by short circuit or short circuit which is the main cause of electric fire. However, in case of electric line, There is a problem that it is difficult to completely prevent the short circuit or short circuit of the electric line.

In addition, recently, it is known that a short circuit is caused by an arc generated on an electric line. Therefore, it is possible to accurately detect and shield an arc caused by various causes such as dielectric breakdown, There is an increasing demand for a fire prevention safety system based on the technology of fire protection.

Therefore, there is a problem that the conventional breaker can not detect an arc caused by a short circuit or a contact failure which is an electric fire cause.

Korean Patent Registration No. 10-1717775 discloses an electric fire analysis and prediction system using an intelligent distribution board for detecting an arc and a risk factor of an electric fire such as a leakage current, an overcurrent, and a short circuit.

However, in the above-mentioned patent, there is a limit to detect a fire only by detecting an arc through monitoring through a distribution board.

Korea registered patent [10-1717775] (Registration date: March 13, 2017) Korea registered patent [10-1583184] (Registration date: December 30, 2015) Korean Patent Publication [10-2016-0149406] (Publication date: December 28, 2016) Korea Registered Patent [10-0993205] (Registered on Nov. 03, 2010)

SUMMARY OF THE INVENTION Accordingly, the present invention has been made keeping in mind the above problems occurring in the prior art, and it is an object of the present invention to provide a disaster detection and control apparatus and an LED illumination unit including a plurality of sensors in a building or each office And analyzing the sensed data collected through the plurality of sensors by using a data mining algorithm and a situation recognition analysis algorithm, it is possible to prevent the occurrence of a disaster in advance, and in the event of a disaster, And can quickly evacuate through the output color, the output light amount or the output pattern, or the sound output of the LED lighting unit according to the disaster training control signal, the disaster prevention control signal or the disaster occurrence control signal, Detection of disasters that can minimize human casualties in the event of a disaster, This includes providing the training function LED lighting system and method.

The objects of the embodiments of the present invention are not limited to the above-mentioned objects, and other objects not mentioned can be clearly understood by those skilled in the art from the following description .

According to an embodiment of the present invention, there is provided an LED lighting system having a disaster prevention detection function and a disaster training function, the system including at least one office (store) in a building or a predetermined area An LED illumination unit 101 for controlling and outputting an output color, an output light amount or an output pattern of the LED according to a disaster prevention control signal, a disaster occurrence control signal, or a disaster training control signal; A disaster detection and control device (110, 130, 140, 150) provided in each office (store); And receiving the sensing data from the disaster detection and control device, analyzing the real-time disaster safety state using a data mining algorithm and a situation recognition analysis algorithm, and transmitting the disaster prevention control signal, the disaster occurrence control signal, And a disaster detection analysis server (120) for transmitting the disaster information to the control device, wherein the disaster detection and control device (110, 130, 140, 150) comprises: a sensor unit (111) having a plurality of sensors; And a controller for generating detection data including an identification number for each office (store), an identification number for each sensor, and a detection signal according to the detection signal transmitted from the sensor unit, and receiving the disaster prevention control signal, A control unit (113) for controlling the output of the LED lighting unit according to a disaster generation control signal, a disaster generation control signal, or a control signal generated by itself; A communication unit 115 for transmitting the detection data to the disaster detection analysis server and receiving the disaster prevention control signal, the disaster occurrence control signal or the disaster response control signal from the disaster detection analysis server; And a storage unit 117 for storing data necessary to control the output of the LED illumination unit according to the disaster prevention control signal, the disaster generation control signal, or the disaster training control signal.

Further, in a disaster prevention advance detection and disaster training method using an LED illumination system according to an embodiment of the present invention, detection data is transmitted from a disaster detection and control device including a plurality of sensors provided in each office (store) A disaster detection analysis step (S610) for detecting and analyzing a disaster occurrence in advance; (S620) for controlling the output of the LED lighting unit according to the disaster prevention control signal and the disaster training control signal when the occurrence of the disaster is predicted and the disaster training is required as a result of the analysis of the disaster detection analysis step (S610) ); And a disaster occurrence control step (S630) of controlling an output of the LED illumination unit according to a disaster occurrence control signal if a disaster occurs as a result of analysis of the disaster detection analysis step (S610) (S710) for collecting and storing sensing data received from the disaster detection and control device provided in each office (store); A data conversion step (S720) of extracting various types of qualities from the collected sensed data and transforming the data; A clustering step (S730) of clustering a plurality of offices (stores) by analyzing the converted data; A community feasibility study step (S740) for evaluating and evaluating the feasibility of the community; A disaster occurrence pattern determination step (S750) of calculating a threshold value of the sensed data for each cluster for which validity has been examined and determining a disaster occurrence pattern according to a threshold value of the sensed data; And a disaster detection step (S760) for predicting a disaster safety state of each office (store) according to the calculated threshold value for each community and transmitting the predicted disaster safety state to the disaster detection and control device.

According to the LED lighting system and method of providing disaster prevention detection and disaster training function according to an embodiment of the present invention, a plurality of sensors installed in each office (store) using a data mining algorithm and a situation recognition analysis algorithm An output light amount or an output pattern of the LED lighting unit or an output pattern of the LED lighting unit according to a disaster training control signal, a disaster prevention control signal, or a disaster occurrence control signal, It is possible to evacuate disaster quickly by making it possible to evacuate quickly through the output, and it can minimize the damage of people in the event of a disaster.

In addition, according to the LED lighting system and method of the present invention having the disaster prevention detection function and the disaster training function according to an embodiment of the present invention, each office (store) is provided with a room sensor, , So if a motion is detected after a late office (store) is locked, it can be considered as a risk factor.

According to an embodiment of the present invention, the disaster detection and control apparatus includes the sensor unit including the volatile organic compound gas sensor, When a volatile organic compound gas having a threshold value or more is detected, damage to the coating of the electric wire is detected, so that the user can be notified of the damage and the electric equipment can be inspected.

Meanwhile, according to the LED lighting system and method in which disaster prevention detection and disaster training functions are provided in accordance with an embodiment of the present invention, based on data sensed by a disaster detection and control device provided in each office (store) It is possible to estimate the disaster occurrence patterns in offices (shops) included in the same community by estimating the similarity of the disaster occurrence patterns, , It is possible to quickly detect whether the disaster safety state of each office (store) is abnormal before a disaster occurs.

In addition, according to the LED lighting system and its method provided with disaster prevention detection and disaster training function according to an embodiment of the present invention, when the possibility of a disaster occurring in a building and a predetermined area is detected, An alarm signal according to the disaster prevention control signal can be outputted through the LED illumination unit or the speaker by transmitting the disaster prevention control signal to an office (store) located within a predetermined radius of the vicinity of the disaster prevention control station (shop).

FIG. 1 is an explanatory view showing a building equipped with an LED lighting system equipped with disaster prevention detection function and disaster training function according to the present invention. FIG.
2 is a block diagram of an embodiment of an LED lighting system equipped with disaster prevention detection and disaster training functions according to the present invention.
3 is a detailed block diagram of an embodiment of a disaster detection and control apparatus according to the present invention.
4 is a detailed configuration diagram of an embodiment of a disaster detection analysis server according to the present invention.
5 is an explanatory diagram of an embodiment of a disaster prevention advance detection method according to the present invention.
FIG. 6 is a flow chart of an embodiment of a disaster prevention detection and disaster training method using the LED lighting system according to the present invention.
7 is a detailed flowchart of the step " S610 "

While the invention is susceptible to various modifications and alternative forms, specific embodiments thereof are shown by way of example in the drawings and will herein be described in detail. It is to be understood, however, that the invention is not to be limited to the specific embodiments, but includes all modifications, equivalents, and alternatives falling within the spirit and scope of the invention.

It is to be understood that when an element is referred to as being "connected" or "connected" to another element, it may be directly connected or connected to the other element, .

On the other hand, when an element is referred to as being "directly connected" or "directly connected" to another element, it should be understood that there are no other elements in between.

The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The singular expressions include plural expressions unless the context clearly dictates otherwise. In the present application, the term " comprises " or " having ", etc. is intended to specify the presence of stated features, integers, steps, operations, elements, parts, or combinations thereof, And does not preclude the presence or addition of one or more other features, integers, integers, steps, operations, elements, components, or combinations thereof.

Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. Terms such as those defined in commonly used dictionaries are to be interpreted as having a meaning consistent with the meaning in the context of the relevant art and are to be construed as ideal or overly formal in meaning unless explicitly defined in the present application Do not.

Hereinafter, the present invention will be described in more detail with reference to the accompanying drawings. Prior to this, terms and words used in the present specification and claims should not be construed as limited to ordinary or dictionary terms, and the inventor should appropriately interpret the concept of the term appropriately in order to describe its own invention in the best way. The present invention should be construed in accordance with the meaning and concept consistent with the technical idea of the present invention. Further, it is to be understood that, unless otherwise defined, technical terms and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. Descriptions of known functions and configurations that may be unnecessarily blurred are omitted. The following drawings are provided by way of example so that those skilled in the art can fully understand the spirit of the present invention. Therefore, the present invention is not limited to the following drawings, but may be embodied in other forms. In addition, like reference numerals designate like elements throughout the specification. It is to be noted that the same elements among the drawings are denoted by the same reference numerals whenever possible.

FIG. 1 is an explanatory view showing a building including an LED lighting system equipped with a disaster prevention detection function and a disaster training function according to the present invention. FIG. 2 is a block diagram of a building equipped with a disaster prevention detection and disaster training function FIG. 3 is a detailed configuration diagram of an embodiment of a disaster detection and control apparatus according to the present invention, FIG. 4 is a detailed configuration diagram of an embodiment of a disaster detection analysis server according to the present invention to be.

As shown in FIGS. 1 to 4, the LED lighting system provided with the disaster prevention detection function and the disaster training function according to the present invention can be implemented by a disaster prevention control signal, a disaster training control signal, An LED illuminating unit 101 for outputting controlled output light, an output light amount or an output pattern, a first disaster detection and control unit 110 connected to a server (disaster detection analysis server 120) through a communication network, And a control device 130, a third disaster detection and control device 140 and an Nth disaster detection and control device 150, and a disaster detection analysis server 120.

The disaster detection analysis server 120 may be installed in a building or a predetermined area through a communication network and stores information such as the entire structure of the building or a predetermined area, a passage, an emergency exit path, It is possible to transmit a signal for controlling the LED illumination unit 101 in accordance with the emergency escape-related information.

The LED lighting system provided with the disaster prevention detection function and the disaster training function according to the present invention includes a user terminal 170 informing the disaster safety state analyzed by the disaster detection analysis server 120, And a disaster management server 180 for transferring information.

Here, the communication network may be configured without regard to its communication mode such as wired and wireless, and may be a local area network (LAN), a metropolitan area network (MAN), a wide area network (WAN) And the like. Preferably, the communication network referred to in the present invention may be the well-known Internet.

Each office in the building 100 or each store within the preset area may have a disaster detection and control device 110, 130, 140, 150 associated with the LED illumination unit 101, 110, 130, 140 and 150 includes a sensor unit 111, a control unit 113, and a communication unit 115. The office (store) provided with the respective disaster detection and control devices 110, 130, 140 and 150 is provided with an electric power distribution panel 102, And emergency power are selectively connected to control the output of the LED lighting unit 101 to be possible.

In this case, the disaster detection and control devices 110, 130, 140, and 150 may transmit signals for controlling the output color, output light amount, or output pattern of the LED illumination unit to the disaster detection analysis server 120, Or from the storage unit 117 provided in the disaster detection and control apparatus and provided through the LED illumination unit 101.

The sensor unit 111 includes a temperature sensor 1111, a volatile organic compound gas sensor 1112, a vibration sensor 1113, a room sensor 1114, a sound sensor 1115, 1116).

The temperature sensing sensor 1111 senses the temperature in each office (store), and the volatile organic compound gas sensing sensor 1112 senses volatile organic compound gas in each office (store) The occupant sensor 1113 senses the vibration of a predetermined size or more, the occupant sensor 1114 senses the movement of the human body, senses the presence or absence of the human body in each office (store) and passage of people between the passages, The sensing sensor 1115 senses sound generation in a predetermined size or more in each office (store), and the electrical load sensing sensor 1116 senses an electrical load in each office (store).

Volatile organic compounds (VOCs) are volatile organic compounds (VOCs) that volatilize in the air and generate odor and ozone. They are carcinogens that cause damage to the nervous system through skin contact and inhalation. Benzene, formaldehyde, toluene, xylene, ethylene, styrene, and acetaldehyde.

These volatile organic compounds often cause bad odors even at low concentrations, and the compounds themselves are directly harmful to the environment and human body, or participate in photochemical reactions in the atmosphere and generate secondary pollutants such as photochemical oxides.

When the heat is generated in the electric wire switchboard, a specific volatile organic compound gas is generated in the insulating cover material. It is possible to detect a possibility of fire by detecting such volatile organic compound gas before a fire occurs.

When heat is generated due to an increase in temperature at the time of electrical fire, volatile organic compounds such as benzene and toluene are generated in the wire covering and the PE material, so that the fire ignition factors can be detected and prevented in advance.

For example, the semiconductor type volatile organic compound gas detection sensor is optimized by a heating substrate designing technique and algorithm of a semiconductor chip type, and is capable of securing stability and reliability.

Accordingly, the temperature detection sensor 1111 and the volatile organic compound gas detection sensor 1112 can detect the occurrence of a fire.

The vibration detection sensor 1113 can be used to detect earthquake or interlayer noise.

The occupant detection sensor 1114 is a sensor used to grasp the movement of the human body in a predetermined space and can control the turn-on time of the LED lighting unit 101 and the brightness of the LED lighting unit 101 according to the detection result.

The operation of the LED lighting unit 101 can also be controlled by using the sensing data of the occupant detection sensor 1114. [

Energy can be saved by turning on the LED lighting unit 101 only for a predetermined time or by controlling the brightness of the LED lighting unit 101 only when sensing the movement of the human body, It may also reduce the risk of occurrence.

The electrical load sensing sensor 1116 includes: It is a sensor that senses when the rated voltage is irregular and the load is over the rated capacity and normal voltage is not supplied.

Processing circuit of the electric circuit breaker and the circuit breaker switch so that the switch of the electric circuit board 102 can be immediately turned off so as not to cause a fire when an event on the electric load occurs. In this case, it is also possible to control the LED lighting unit 101 to supply emergency power.

The controller 113 controls the electric power distribution board 102 according to the detection signal received from the sensor unit 111 and the disaster prevention control signal received from the disaster analysis server 120, And the output of the LED lighting unit 101 can be controlled.

The control unit 113 transmits the sensing data including the sensing signal received from the sensor unit 111, the identification number of the office (shop) and the sensor type to the communication unit 115, The disaster prevention control signal or the disaster occurrence control signal transmitted from the disaster analysis server 120 to the server 120 and outputs the output color of the LED illumination unit 101, And an output pattern can be controlled.

The control unit 113 may control the output of the LED lighting unit 101 according to the disaster detection training control signal stored in the storage unit 117 according to the sensing signal received from the sensor unit 111 have.

Each of the disaster detection and control devices 110, 130, 140, and 150 may include a control signal for lighting and dimming control of the LED lighting unit 101, a disaster prevention control signal, a disaster training control signal, A signal for controlling the output color, the output light quantity and the output pattern of the LED lighting unit 101, and an identification number of the office (shop).

At this time, the controller 113 performs an OR operation to satisfy the disaster prevention control signal, the disaster training control signal, the disaster occurrence control signal, or the self control signal received from the disaster analysis server 120, The lighting unit 101 can be controlled.

Under the control of the controller 113, the communication unit 115 transmits sensed data including the sensing signal of the sensor unit 111, the identification number (ID) of the office (shop) and the sensor type, Analysis server 120 and receives the disaster prevention control signal, the disaster training control signal or the disaster occurrence control signal from the disaster detection analysis server 120 and transmits the disaster prevention control signal or the disaster occurrence control signal to the control unit 113.

Meanwhile, the control unit 113 may output a voice signal through the speaker 119 according to the disaster prevention control signal, the disaster training control signal, the disaster occurrence control signal, and the self control signal.

Each of the disaster detection and control devices 110, 130, 140, and 150 includes an electric distribution board 102 for supplying power to the respective components. Each of the disaster detection and control devices 110, 130, 140, and 150 may be powered by a solar cell module (not shown). When the disaster detection and control devices 110, 130, 140, and 150 are powered by the solar cell module, the power supply to the electric distribution panel 102 may be cut off or may be driven during a power failure. There is a merit that fire detection can be done even in a situation.

At least a part of the sensor unit 111, the control unit 113, the communication unit 115 and the storage unit 117 are program modules communicating with the respective disaster detection and control devices 110, 130, 140 and 150 . These program modules may be included in the disaster detection and control devices 110, 130, 140 and 150 in the form of an operating system, application program modules and other program modules, and may be stored physically on various known storage devices . These program modules may also be stored in a remote storage device capable of communicating with each of the disaster detection and control devices 110, 130, 140, These program modules include, but are not limited to, routines, subroutines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular abstract data types as described below in accordance with the present invention.

The communication unit 115 may include one or more modules that enable wire / wireless communication with the disaster detection analysis server 120. For example, the communication unit 115 may include a wired / wireless Internet module (not shown), a short-range communication module (not shown), and a location information module (not shown).

The wired / wireless Internet module refers to a module for wired / wireless Internet access, and it can be built in or externally installed in a street light. WLAN (Wi-Fi), Wibro (Wireless broadband), Wimax (World Interoperability for Microwave Access), WCDMA (Wideband Code Division Multiple Access), HSPA (High Speed Packet Access) LTE (Long Term Evolution) or the like can be used.

The short-range communication module is a module for short-range communication. Bluetooth, Radio Frequency Identification (RFID), Infrared Data Association (IrDA), Ultra Wideband (UWB), ZigBee, LoRa, etc. may be used as the technique of short range communication.

The location information module is a module for acquiring the location of a corresponding office (store), and a representative example thereof is a global positioning system (GPS) module. According to the present technology, the GPS module calculates distance information and accurate time information from three or more satellites, and then applies trigonometry to the calculated information to obtain three-dimensional current position information according to latitude, longitude, and altitude It can be calculated accurately. At present, a method of calculating position and time information using three satellites and correcting an error of the calculated position and time information using another satellite is widely used. In addition, the GPS module can calculate speed information by continuously calculating the current position in real time.

The disaster detection analysis server 120 receives the detection data from the disaster detection and control devices 110, 130, 140 and 150, analyzes the data using a data mining algorithm and a situation recognition analysis algorithm, And transmits a training control signal or a disaster occurrence control signal to the disaster detection and control device 110, 130, 140,

That is, the disaster detection analysis server 120 according to the present invention can predict and inform a situation in which a disaster occurs by using a data mining algorithm and a situation recognition analysis algorithm of the Big Data.

Big data means a vast amount of information digitized. It can filter out unnecessary data from big data and extract and analyze only useful information to read people's thoughts, opinions and trends and to predict their behavior in advance. Big Data is one of the next generation information technology (IT) technology that is not only in Korea but also in the world because of its usefulness.

The domestic big data market is expected to reach 300 billion won in 2015 and grow to 1 trillion won in 2020. The domestic market related to Big Data is growing by more than 28% every year. Although the use of Big Data is concentrated in research and consulting, it is likely to be used in commercial and advertising markets.

Big data refers to data sets beyond the ability to collect, store, manage and analyze existing data. Big data can be classified into stereotyped data, semi-stereotyped data, and non-stereotyped data according to the degree of stereotyping.

Structured data refers to data stored in fixed fields. In other words, data is stored in a certain format. Semi-structured data refers to data that is not stored in a fixed field but contains metadata or schema. Examples of semi-structured data include Extensible Mark-up Language (XML) and Hypertext Mark-up Language (HTML). Unstructured data refers to data that is not stored in a fixed field. Examples of the unstructured data include text documents, image data, moving image data, and audio data.

Big data analysis techniques include text mining, data mining, opinion mining, social network analysis, cluster analysis, neural network analysis, and Markov model (markov model), but is not limited to the illustrated analytical techniques.

Text mining is a technique for extracting and processing useful information based on natural language processing techniques in semi-structured text data or unstructured text data. Natural Language Processing (NLP) technology is a technology that enables natural language understanding and natural language generation. Natural language is a term used by people to communicate and is the opposite of artificial language (computer language). Natural language understanding is the natural interpretation of the natural language to make it understandable to the computer. Natural Language Generation refers to the ability of a computer to output natural language. Natural language processing techniques consist of morpheme analysis, parsing, semantic analysis, and phonetic analysis.

Data mining is the process of finding the knowledge of interest in a database by any method (sequential pattern, similarity, etc.). In other words, data mining is a process of discovering useful information in a large amount of data, and it means a technology that can find not only the expected information but also the unexpected information. Data mining can be used to determine the relevance of information and to maximize profits by creating valuable information and applying it to decision making. Hidden knowledge, unexpected trends, or new rules based on all available source data, including daily transaction data, customer data, product data or customer response data from various marketing activities, as well as other external data And use it as information for actual business decision making.

Reputation analysis can be applied to social media such as blogs, social network services (SNS), wikis, UCCs, micro-blogs, It is a technology that collects and analyzes texts and determines the reputation (for example, affirmative, negative, neutral) of a product or service.

Social network analysis is a technique for analyzing and extracting user influence, interest, and propensity based on social network connection structure and connection strength.

Cluster analysis is a technique for collecting groups with similar characteristics, finally combining groups with similar characteristics.

In the present invention, a clustering technique for data mining is used.

The disaster detection analysis server 120 includes a transceiver 121, a collecting and storing unit 122, a data converting unit 123, a clustering unit 124, a cluster attaching unit 125, a situation recognizing unit 126, And a determination and control unit 127. [

The transceiver unit 121 receives the sensed data from the disaster detection and control devices 110, 130, 140 and 150 and transmits the disaster prevention control signal disaster training control signal or the disaster occurrence control signal .

The collection and storage unit 122 collects and stores the disaster detection data received from the disaster detection and control apparatuses 110, 130, 140 and 150.

The sensed data collected and stored may include the source of the sensed data (office (store)), the type of sensor, the collected time, raw data of the sensor, and status information.

The data conversion unit 123 extracts various types of characteristics (characteristics, features, and features) from the collected sensed data and converts the extracted data.

Feature extraction refers to transforming the input data into a reduced representation set (or feature vector) of features if the input data is too large to process and there is a doubt that there is a significant unnecessary duplication. That is, the process of converting input data into a set of features is called feature extraction or feature extraction. Feature extraction is performed on the raw data before applying the k-NN algorithm to the transformed data into the feature space.

The data conversion unit 123 extracts the qualities including the minimum value, the maximum value, the average, the variance, or the standard deviation from the sensed data using the window of the predetermined size and whether or not the overlap is detected. The extracted m qualities are defined as m-dimensional feature vectors for each office (store).

The clustering unit 124 analyzes the converted data (feature vectors) to cluster a plurality of offices (stores).

For example, in the present invention, a K-means algorithm is used for clustering. In another embodiment, clustering may be accomplished using techniques such as Euclidean distance and cosine similarity.

The K-means algorithm is an algorithm for grouping the given data into k clusters. The K-means algorithm works by minimizing the variance of the distance difference with each cluster.

The K-means algorithm belongs to the partitioning method among the clustering methods. The partitioning method divides the given data into several partitions (groups). For example, assume that n data objects have been input. In this case, the partitioning method divides the input data into k groups smaller than or equal to n, where each cluster forms a cluster. In other words, it divides the data into k groups of one or more data objects. At this time, the process of dividing the group is performed in such a manner as to minimize the cost function such as the distance-based dissimilarity between the groups. In this process, the similarity between the data objects in the same group is increased, The degree of similarity with the data object in the area is reduced. The K-average algorithm determines the sum of squares of the distance between the centroid of each group and the data object in the group as a cost function and performs clustering by updating the belonging group of each data object in the direction of minimizing the function value do.

That is, given a set of n m-dimensional data objects x, the K-means algorithm sets n data objects to k (<= n) sets S = {S ₁ , S ₂ , ... , S _k }. In other words, if μi is the center of the i-th cluster and S _i is a set of points belonging to the i-th cluster, finding the set S that minimizes the sum of squares of the distances between the objects in the center- .

That is, a plurality of offices (stores) are clustered using a K-average algorithm satisfying the following Equation (1).

&Quot; (1) &quot;

n is the number of an office or shop, k is the number of clusters (cluster), x is an m- dimensional of each office or store feature vector, μ _i has the form of m- dimension to the center of the i-th cluster vector, S _i is a set of offices or stores belonging to the i-th cluster.

The clustering unit 124 receives the detection data of the newly added disaster detection and control devices 110, 130, 140 and 150 and transmits the detection data to the corresponding office (store) using a k-NN (k-nearest neighbor) Can be determined to which cluster.

In the k-NN (k-nearest neighbors) algorithm, for example, Euclidean distances may be used.

The Euclidean distance between the center point (center) included in each cluster and the feature vector of the newly added office (store) can be calculated.

Since each office (store) has an m-dimensional feature vector, two vectors of m-dimension p = (p ₁ , p ₂ , ..., p _m ) and q = (q ₁ , q ₂ , ... ., the Euclidean distance between the q _m) (L _u) is calculated using the following <equation 2>.

&Quot; (2) &quot;

Where q is the m-dimensional feature vector of the newly added office or store, and j is an integer from 1 to m, where m is the number of dimensions of the feature vector, p is the m- It is a constant. p and q may be mutually exclusive.

Therefore, after computing i Euclidean distances for i clusters, closest clusters can be defined as a cluster of newly added offices (stores).

On the other hand, after calculating the Euclidean distance between the feature vector included in each cluster and the feature vector of the newly added office (store), the average value is calculated and the cluster having the smallest average value is added to the newly added office ). &Lt; / RTI &gt;

The average value (m _Lu ) of the Euclidean distance of each cluster can be expressed by Equation (3) below.

&Quot; (3) &quot;

Here, m is the number of dimensions of the feature vector, a is the number of offices or stores included in an arbitrary cluster, r _t is m-dimensional feature vectors included in an arbitrary cluster, q is m - dimensional feature vector, t is a constant with an integer value from 1 to a, and j is a constant with an integer value from 1 to m. Therefore, after calculating the average value of the Euclidean distances by the calculation of Equation (3) for each of the k clusters, the clusters having the smallest average value may be defined as clusters of newly added offices or stores.

The cluster evaluation unit 125 evaluates the validity of the clusters clustered by the clustering unit 124 and examines whether the clusters are properly formed.

The community evaluation unit 125 evaluates the validity of the community through the silhouette index s (i) expressed in Equation (4) below.

&Quot; (4) &quot;

Where a (i) is the average of the distances between elements in the cluster to which the i-th entity belongs and b (i) is the distance between elements in the i-th entity and the cluster to which the i-th entity does not belong And each of the elements belonging to an individual or a cluster is an m-dimensional feature vector, respectively.

In the most ideal case a (i) is zero. In other words, this is the case where all the individuals in a community are not separated. The silhouette index is then 1. In the worst case, b (i) is zero. In other words, the different clusters are not distinguished at all, and the silhouette index is -1 at this time. If the average silhouette index is greater than 0.5, the cluster results are considered valid.

Meanwhile, the cluster evaluation unit 125 may evaluate the validity of the cluster by calculating a Dunn Index I (C) expressed in Equation (5).

Equation (5)

Here, k is the number of clusters, C _i is the m-dimensional vector shape as the center point of the i-th cluster, and d _c (C _i , C _j ) is the distance between C _i and C _j .

The Dune index is a numerator with the smallest value of the distance between the clusters and a denominator with the maximum value of the distance between the elements in the cluster. In other words, the closer the distance between the clusters, the smaller the dispersion within the clusters, the better the clustering result. That is, if the clustering is well done, the Dune index has a large value.

The situation recognition unit 126 calculates a threshold value of the sensed data for each cluster for which the validity is evaluated and determines a disaster occurrence pattern according to the threshold value of the sensed data.

The judgment and control unit 127 predicts a disaster safety state (possibility of a disaster) of each office (store) according to the calculated threshold value for each community and outputs a disaster prevention control signal or a disaster occurrence control signal to the corresponding disaster detection and control unit (110, 130, 140, 150).

The disaster safety state is classified into one of a safety state period that is smaller than the first estimated value, a state state region that is larger than the first estimated value and smaller than the second estimated value, or a dangerous state region that exceeds the second estimated value. (The first estimated value < the second estimated value) Here, the first estimated value and the second estimated value are calculated by numerically calculating the risk factors in various ways according to the disaster occurrence pattern.

The sensed data collected and stored includes the source of the sensed data (office (store)), the type of sensor, the collected time, raw data of the sensor, and status information.

The disaster prevention control signal, the disaster training control signal, or the disaster occurrence control signal may include identification information of the disaster detection and control devices 110, 130, 140, and 150 to be controlled, on-off signals of the LED illumination unit 101, A brightness control signal, and a color control signal.

Meanwhile, the disaster detection analysis server 120 includes a power supply unit (not shown) for supplying power to each component.

5 is an explanatory view of an embodiment of a disaster detection analysis method according to the present invention.

FIG. 5 is a time-based graph showing sensed data received from the disaster detection and control apparatus 110 by the disaster detection analysis server 120 according to the present invention.

For example, for a certain community, if the threshold value for the detection signal of the volatile organic compound gas detection sensor 1112 is the threshold value A and the threshold value for the detection signal of the vibration detection sensor 1113 or the sound detection sensor 1115 It is assumed that the threshold value is the threshold value B, the threshold value for the sensing signal of the temperature sensing sensor 1111 is the threshold value C, and the threshold value for the sensing signal of the electric load sensing sensor 1116 is set to the threshold value D .

The disaster detection analysis server 120 detects vibration or sound as the detection signal of the vibration detection sensor 1113 or the sound detection sensor 1115 exceeds the threshold value B 401. [ It is possible to detect a sudden loud sound in the office (shop). In this case, the disaster detection analysis server 120 or the control unit 113 can output a warning sound indicating that a loud sound or vibration is detected in the corresponding office (store).

On the other hand, the disaster detection analysis server 120 can determine that there is no risk of a disaster (e.g., fire) before the " first detection " The detection signal of the electric load detecting sensor 1116 has been " first detected " by exceeding the threshold value D (505).

At this time, since the detection signal of the temperature sensor 1111 exceeds the threshold value C (503) but the detection signal of the electric load sensor 1116 does not exceed the threshold value A, the possibility of a fire is low And it is possible to consider various situations such as the increase in the number of people in the office (store) or the use of the heater.

Then, over time, the detection signal of the volatile organic compound gas detection sensor 1116 is shown as being " secondary sensed " by exceeding the threshold value A (502). Since the detection signal of the electric load detecting sensor 1116 at the time of the " second detection " exceeds the threshold value D (506), it can be considered as a risk of fire occurrence.

For example, it can be judged as a safety state until "first detection", and it can be judged as a caution state before the "second detection" state, and it can be judged as a dangerous state after "second detection".

Situation recognition can be classified into various types according to types of disasters including the sensing signals and the duration of the sensor units. In addition, new disaster patterns can be continuously updated.

FIG. 6 is a flow chart of an embodiment of a pre-disaster detection and disaster training method using the LED lighting system according to the present invention.

(110, 130, 140, 150) including a plurality of sensors provided in each office (store) in a building or within a predetermined area to detect and analyze the occurrence of a disaster in advance (S610).

If the disaster is predicted to occur and the disaster training is required as a result of the analysis of the disaster detection analysis step S610, the output of the LED illumination unit 101 is controlled according to the disaster prevention control signal and the disaster training control signal at step S620.

If a disaster occurs, the output of the LED lighting unit 101 is controlled according to a disaster occurrence control signal in operation S630.

Of course, when the disaster detection analysis server 120 receives a disaster related event from the external disaster management server 180, it can directly transmit the disaster occurrence control signal to the disaster detection and control apparatuses 110, 130, 140 and 150 have.

Meanwhile, when the disaster detection and control device 110, 130, 140, or 150 itself detects a disaster, the output of the LED illumination unit 101 may be controlled according to a disaster occurrence control signal.

The disaster prevention control signal, the disaster training control signal, or the disaster occurrence control signal may include an on-off signal, a dimming control signal, a brightness control signal, and a color control signal of the LED lighting unit 101.

Here, the first illumination unit control step of controlling the LED illumination unit 101 according to the disaster prevention control signal, the disaster training control signal or the disaster occurrence control signal received from the disaster analysis server 120, The second illuminating unit controlling step of controlling the illuminating unit 101 may be ORed to control the LED illuminating unit 101.

The disaster detection and control apparatuses 110, 130, 140, and 150 may control the disconnection of power according to the disaster occurrence control signal. The control generation control signal may be transmitted to all disaster detection and control devices of offices (shops) included in the preset area including the corresponding office (store).

7 is a detailed flowchart of the step " S610 " in Fig.

First, the collection and storage unit 122 of the disaster detection analysis server 120 receives the detection data from the disaster detection and control apparatus 110 provided in each office (store) through the transmission and reception unit 121, (S710).

The sensed data collected and stored includes the source of the sensed data (office (store)), the type of sensor, the collected time, raw data of the sensor, and status information.

The data conversion unit 123 of the disaster detection analysis server 120 extracts various types of qualities from the sensed data to be collected and converts the extracted data (S720).

In the feature extraction method, features including a minimum value, a maximum value, an average, a variance, or a standard deviation are extracted from the sensed data using a window of a predetermined size and overlap or not. The extracted m qualities are defined as m-dimensional feature vectors for each office (store).

The clustering unit 124 of the disaster detection analysis server 120 analyzes the converted data to cluster a plurality of offices (stores) (S730).

The clustering unit 124 clusters a plurality of offices (stores) using a K-average algorithm satisfying Equation (6).

&Quot; (6) &quot;

Here, given a set of n m-dimensional data objects x, the K-means algorithm sets n data objects to k (<= n) sets S = S ₁ , S ₂ , ... , S _k , where n is the number of offices or stores, k is the number of clusters, x is the m-dimensional feature vector of each office or store, μ _i is the center of the i- Dimensional vector, and S _i is a set of offices or stores belonging to the i-th cluster.

The clustering unit 124 receives the detection data of the newly added disaster detection and control devices 110, 130, 140 and 150 and transmits the detection data to the corresponding office (store) using a k-NN (k-nearest neighbor) Can be determined to which cluster.

In the k-NN (k-nearest neighbors) algorithm, for example, Euclidean distances may be used.

The Euclidean distance between the center point (center) included in each cluster and the feature vector of the newly added office (store) can be calculated.

Since each office (store) has an m-dimensional feature vector, two vectors of m-dimension p = (p ₁ , p ₂ , ..., p _m ) and q = (q ₁ , q ₂ , ... ., Euclidean distance (L _u between the q _m)) is calculated using the following <equation 7>.

&Quot; (7) &quot;

Where q is the m-dimensional feature vector of the newly added office or store, and j is an integer from 1 to m, where m is the number of dimensions of the feature vector, p is the m- It is a constant. p and q may be mutually exclusive.

Therefore, after computing i Euclidean distances for each of the i clusters, a closest cluster (cluster) can be defined as a cluster of newly added offices (stores).

On the other hand, after calculating the Euclidean distance between the feature vector included in each cluster and the feature vector of the newly added office (store), the average value is calculated and the cluster having the smallest average value is added to the newly added office ). &Lt; / RTI &gt;

The average value (m _Lu ) of the Euclidean distance of each cluster can be expressed by Equation (8) below.

&Quot; (8) &quot;

Here, m is the number of dimensions of the feature vector, a is the number of offices or stores included in an arbitrary cluster, r _t is m-dimensional feature vectors included in an arbitrary cluster, q is m - dimensional feature vector, t is a constant with an integer value from 1 to a, and j is a constant with an integer value from 1 to m. Accordingly, after calculating the average value of the Euclidean distances by the calculation of Equation (8) for each of the k clusters, the clusters having the smallest average value may be defined as clusters of newly added offices or stores.

The community evaluation unit 125 of the disaster detection analysis server 120 evaluates the validity of the clusters clustered by the clustering unit 124 and examines whether the clusters are properly established (S740).

The community evaluation unit 125 evaluates the validity of the community through the silhouette indicator s (i) expressed in Equation (9).

&Quot; (9) &quot;

Where a (i) is the average of the distances between elements in the cluster to which the i-th entity belongs and b (i) is the distance between elements in the i-th entity and the cluster to which the i-th entity does not belong And each of the elements belonging to an individual or a cluster is an m-dimensional feature vector, respectively.

In the most ideal case a (i) is zero. In other words, this is the case where all the individuals in a community are not separated. The silhouette index is then 1. In the worst case, b (i) is zero. In other words, the different clusters are not distinguished at all, and the silhouette index is -1 at this time. If the average silhouette index is greater than 0.5, the cluster results are considered valid.

Meanwhile, the community evaluation unit 125 of the disaster detection analysis server 120 may evaluate the validity of the community by calculating a Dunn Index (I (C)) expressed in Equation (10) .

&Quot; (10) &quot;

Here, k is the number of clusters, C _i is the m-dimensional vector shape as the center point of the i-th cluster, and d _c (C _i , C _j ) is the distance between C _i and C _j .

The Dune index is a numerator with the smallest value of the distance between the clusters and a denominator with the maximum value of the distance between the elements in the cluster. In other words, the closer the distance between the clusters, the smaller the dispersion within the clusters, the better the clustering result. That is, if the clustering is well done, the Dune index has a large value.

Thereafter, the situation recognition unit 126 of the disaster detection analysis server 120 calculates a threshold value of the sensed data for each cluster for which the validity is examined, and determines a disaster occurrence pattern according to the threshold value of the sensed data (S750) .

A disaster prevention control signal, a disaster training control signal, or a disaster occurrence control signal to the disaster detection and control apparatuses 110, 130, 140, and 150 of the office (store) by predicting the disaster safety state of each office (store) (S760).

In addition, the disaster safety state corresponding to the disaster prevention control signal, the disaster training control signal, or the disaster occurrence control signal is notified through the user terminal 170 (S770).

The disaster safety state is classified into one of a safety state period that is smaller than the first estimated value, a state state region that is larger than the first estimated value and smaller than the second estimated value, or a dangerous state region that exceeds the second estimated value. (The first estimated value < the second estimated value) Here, the first estimated value and the second estimated value are calculated by numerically calculating the risk factors in various ways according to the disaster occurrence pattern.

The disaster detection analysis server 120 updates the cluster, the threshold, and the disaster occurrence pattern based on the detection data collected during the predetermined period (S780).

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed exemplary embodiments, Of course, it is also possible to implement a program stored in a computer-readable recording medium for realizing a disaster prevention detection and disaster training method using a recording medium and an LED lighting system.

That is, it should be understood by those skilled in the art that disaster prevention detection and disaster training methods using the LED lighting system described above may be provided in a recording medium readable by a computer by tangibly embodying a program of instructions for implementing the same It will be easy to understand. In other words, it can be implemented in the form of a program command that can be executed through various computer means, and can be recorded on a computer-readable recording medium. The computer-readable recording medium may include program commands, data files, data structures, and the like, alone or in combination. The program instructions recorded on the computer-readable recording medium may be those specially designed and configured for the present invention or may be those known and available to those skilled in the computer software. Examples of the computer-readable medium include magnetic media such as hard disks, floppy disks and magnetic tape, optical media such as CD-ROMs and DVDs, and optical disks such as floppy disks. Magneto-optical media and hardware devices specifically configured to store and execute program instructions such as ROM, RAM, flash memory, USB memory, and the like. The computer-readable recording medium may be a transmission medium such as a light or metal line, a wave guide, or the like, including a carrier wave for transmitting a signal designating a program command, a data structure, and the like. Examples of program instructions include machine language code such as those produced by a compiler, as well as high-level language code that can be executed by a computer using an interpreter or the like. The hardware device may be configured to operate as one or more software modules to perform the operations of the present invention, and vice versa.

It will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.

101: LED lighting part 102: electric power distribution board
110: first disaster detection and control device 120: disaster detection analysis server
130: second disaster detection and control device 140: third disaster detection and control device
150: Nth disaster detection and control device 170: User terminal
180: Disaster management server 111: Sensor unit
113: control unit 115:
117: storage unit 119: speaker
1111: Temperature sensor 1112: Volatile organic compound gas sensor
1113: Vibration sensor 1114: Vibration sensor
1115: sound detection sensor 1116: electric load detection sensor
121: Transmitting / receiving unit 122: Collecting and storing unit
123: data conversion unit 124: clustering unit
125: cluster evaluation unit 126: situation recognition unit
127: Judgment and control unit

Claims (10)

1. An LED lighting system having a disaster prevention detection function and a disaster training function,
An LED lighting unit (not shown) provided in the building or one or more offices or shops in a predetermined area and controlling the output color, output light amount or output pattern of the LED according to the disaster prevention control signal, the disaster occurrence control signal, 101);
A disaster detection and control device (110, 130, 140, 150) provided in each of the offices or shops; And
Receiving discovery data from the disaster detection and control device, analyzing a real-time disaster safety state using a data mining algorithm and a situation recognition analysis algorithm, and analyzing a disaster prevention control signal, a disaster occurrence control signal, A disaster detection analysis server 120 for transmitting to the control device;
Lt; / RTI &gt;
The disaster detection and control device (110, 130, 140, 150)
A sensor unit 111 having a plurality of sensors;
And generates sensing data including an identification number for each office or a store, an identification number for each sensor, and a sensing signal according to a sensing signal transmitted from the sensor unit, and transmits the disaster prevention control signal received from the disaster sensing analysis server, A control unit 113 for controlling an output of the LED lighting unit according to a control signal, a disaster training control signal, or a control signal generated by itself; And
A communication unit 115 for transmitting the detection data to the disaster detection analysis server and receiving the disaster prevention control signal, the disaster occurrence control signal or the disaster response control signal from the disaster detection analysis server; And
A storage unit 117 for storing data necessary to control the output of the LED illumination unit according to the disaster prevention control signal, the disaster occurrence control signal,
Lt; / RTI &gt;
The disaster detection analysis server (120)
A transmission / reception unit 121 for receiving detection data from the disaster detection and control device and for transmitting the disaster prevention control signal, a disaster occurrence control signal or a disaster training control signal to the disaster detection and control device;
A collection and storage unit (122) for collecting and storing the sensing data received from the disaster detection and control device;
A data conversion unit 123 for extracting various types of qualities from the collected sensed data and converting the data;
A clustering unit 124 for clustering a plurality of offices or stores by analyzing the converted data;
A cluster evaluating unit 125 for evaluating the validity of the cluster generated by the clustering unit;
A situation recognition unit 126 for calculating a threshold value of the sensed data for each cluster for which the validity is evaluated and determining a disaster occurrence pattern according to a threshold value of the sensed data; And
A disaster prevention control signal, a disaster occurrence control signal, or a disaster training control signal is predicted to be transmitted to the corresponding disaster detection and control apparatus by predicting the disaster safety state of each office or shop according to the calculated threshold value of the community- The determination and control unit 127,
Wherein the LED illumination system includes a disaster prevention detection function and a disaster training function.
The method according to claim 1,
The sensor unit (111)
A temperature sensing sensor 1111 for sensing the temperature in each office or shop;
A volatile organic compound gas detection sensor 112 for detecting volatile organic compound gas in each of the offices or stores;
A vibration detection sensor 1113 for detecting a vibration of a predetermined size or more;
An occupant detection sensor 1114 for detecting the presence of a human body in each of the offices or shops by sensing the movement of the human body;
A sound detection sensor (1115) for detecting sound generation in a predetermined size or more in each of the offices or shops; And
An electrical load sensing sensor 1116 for sensing an electrical load in each office or store,
Wherein the LED illumination system includes a disaster prevention detection function and a disaster training function.
3. The method of claim 2,
The LED lighting unit and the disaster detection and control device are provided in each office or shop or the building to supply electric power to each component of the LED illumination unit and the disaster detection and control unit, and selectively connect normal power and emergency power according to detection of disaster occurrence, And an electric power distribution board (102) for controlling the output of the electric load sensor to be turned off when the electric load event is detected in the electric load detection sensor;
A user terminal 170 for reporting the disaster safety state analyzed by the disaster detection analysis server;
A disaster management server 180 for delivering real-time disaster information through a communication network; And
A speaker 119 for outputting a voice signal in accordance with the disaster prevention control signal, the disaster training control signal, the disaster occurrence control signal, or the control signal of the control unit,
Wherein the LED lighting system includes a disaster prevention detection function and a disaster training function.
delete The method according to claim 1,
Wherein the data conversion unit comprises:
Dimensional feature vector for each office or store, and the extracted m qualities are defined as m-
The clustering unit,
Using a K-means algorithm that satisfies Equation (11) below,
Equation (11)

(Here, given a set of n m-dimensional data objects x, the K-means algorithm sets n data objects to k (<= n) sets S = S ₁ , S _2, ..., by dividing the S _k, n is the number of an office or shop, k is the number of clusters (cluster), x is the dimension of each m- office or store feature vector, μ _i is the i-th cluster Dimensional vector at the center of the i-th cluster, and S _i is the set of offices or stores belonging to the i-th cluster)
The clustering unit,
And a k-NN (k-nearest neighbors) algorithm is used to receive the detection data of the newly added disaster detection and control apparatus and determine to which cluster the corresponding office or store corresponds.
The clustering unit,
the k closest clusters are determined as the cluster of the newly added office or shop after calculating k euclidean distances with respect to the center of k clusters and the feature vector of the newly added office or store, and,
Since each office or store has an m-dimensional feature vector, two vectors of m-dimension p = (p ₁ , p ₂ , ..., p _m ) and q = (q ₁ , q ₂ , ... , characterized in that the calculation using the q _m) to the Euclidean distance (L _u) between the <equation 12>, and
&Quot; (12) &quot;

Where m is the number of dimensions of the feature vector, p is the m-dimensional vector shape at the center of each cluster, q is the m-dimensional feature vector of the newly added office or store, j is an integer value from 1 to m Lt; / RTI &gt;
The clustering unit,
After computing the Euclidean distance between the feature vectors included in each k clusters and the feature vectors of the newly added office or store, the average value is calculated and the clusters having the smallest average value are added to the newly added office or store And a cluster,
The average value (m _Lu ) of the Euclidean distance of each cluster is calculated using the following Equation (13)
&Quot; (13) &quot;

(Where m is the number of dimensions of the feature vector, a is the number of offices or stores included in any cluster, r _t is the m-dimensional feature vectors included in any cluster, and q is the number of offices or stores m-dimensional feature vector, t is a constant having an integer value from 1 to a, and j is a constant having an integer value from 1 to m)
The clustering unit,
Wherein a cluster of a newly added office or shop is determined using Equation (12) or Equation (13).
6. The method of claim 5,
The cluster evaluation unit calculates,
Wherein the silhouette indicator (s (i)) is calculated using Equation (14), and the validity of the cluster is evaluated.
&Quot; (14) &quot;

Where a (i) is the average of the distances between the i-th entity and the elements in the cluster to which the i-th entity belongs, and b (i) And the elements belonging to the individual or cluster are m-dimensional feature vectors, respectively)
In a disaster prevention detection and disaster training method using an LED lighting system,
A disaster detection analysis step (S610) for detecting and analyzing a disaster occurrence in advance by receiving detection data from a disaster detection and control apparatus including a plurality of sensors provided in each office or shop;
(S620) for controlling the output of the LED lighting unit according to the disaster prevention control signal and the disaster training control signal when the occurrence of the disaster is predicted and the disaster training is required as a result of the analysis of the disaster detection analysis step (S610) ; And
(S630) for controlling an output of the LED illuminating unit according to a disaster occurrence control signal when a disaster occurs as a result of the analysis of the disaster detection analysis step (S610)
Lt; / RTI &gt;
The disaster detection analysis step (S610)
A data collecting step (S710) of receiving and collecting sensed data from the disaster detecting and controlling device provided in each office or shop;
A data conversion step (S720) of extracting various types of qualities from the collected sensed data and transforming the data;
A clustering step (S730) of clustering a plurality of offices or stores by analyzing the converted data;
A community feasibility study step (S740) for evaluating and evaluating the feasibility of the community;
A disaster occurrence pattern determination step (S750) of calculating a threshold value of the sensed data for each cluster for which validity has been examined and determining a disaster occurrence pattern according to a threshold value of the sensed data;
A disaster detection step of predicting a disaster safety state of each office or shop according to the calculated threshold value of each community, and transmitting the disaster prevention control signal, the disaster occurrence control signal or the disaster training control signal to the corresponding disaster detection and control device (S760)
And a method for disaster prevention detection and disaster training using the LED lighting system.
8. The method of claim 7,
Wherein the data conversion step comprises:
Dimensional feature vector for each office or store, and the extracted m qualities are defined as m-
In the clustering step,
Average algorithm that satisfies Equation (15) below is used,
&Quot; (15) &quot;

(Here, given a set of n m-dimensional data objects x, the K-means algorithm sets n data objects to k (<= n) sets S = S ₁ , S _2, ..., by dividing the S _k, n is the number of an office or shop, k is the number, x in the cluster (cluster) m- is a dimension of each feature vector an office or shop, μi is the i-th cluster Dimensional vector, S _i is the set of offices or stores belonging to the i-th cluster)
The disaster detection analysis step (S610)
A cluster decision step of receiving the detection data of the newly added disaster detection and control device and deciding which cluster corresponds to the received data
Further comprising:
The cluster determination step may include:
the k closest clusters are determined as the cluster of the newly added office or shop after calculating k euclidean distances with respect to the center of k clusters and the feature vector of the newly added office or store, and,
Since each office or store has an m-dimensional feature vector, two vectors of m-dimension p = (p ₁ , p ₂ , ..., p _m ) and q = (q ₁ , q ₂ , ... , characterized in that calculated using the Euclidean distance (L _u) to the <equation 16> between the q _m) and,
&Quot; (16) &quot;

(Where m is the number of dimensions of the feature vector, p is the center of each cluster, q is the m-dimensional feature vector of the newly added office or store, and j is a constant having an integer value from 1 to m)
The cluster determination step may include:
After computing the Euclidean distance between the feature vectors included in each k clusters and the feature vectors of the newly added office or store, the average value is calculated and the clusters having the smallest average value are added to the newly added office or store And a cluster,
The average value (m _Lu ) of the Euclidean distance of each cluster is calculated by using the following equation (17)
&Quot; (17) &quot;

(Where m is the number of dimensions of the feature vector, a is the number of offices or stores included in any cluster, r _t is the m-dimensional feature vectors included in any cluster, and q is the number of offices or stores m-dimensional feature vector, t is a constant having an integer value from 1 to a, and j is a constant having an integer value from 1 to m)
The cluster determination step may include:
A cluster of a newly added office or shop is determined by using Equation (16) or Equation (17).
9. The method of claim 8,
The community feasibility examination step includes:
Calculating a silhouette indicator (s (i)) using Equation (18), and evaluating the validity of the cluster; and detecting the disaster occurrence and disaster training method using the LED illumination system.
&Quot; (18) &quot;

Where a (i) is the average of the distances between the i-th entity and the elements in the cluster to which the i-th entity belongs, and b (i) And the elements belonging to the individual or cluster are m-dimensional feature vectors, respectively)
8. The method of claim 7,
The disaster detection analysis step (S610)
A state notification step (S770) of informing the user terminal of the disaster safety state corresponding to the disaster prevention control signal, the disaster training control signal or the disaster occurrence control signal; And
An update step (S780) of updating the cluster, the threshold value, and the disaster occurrence pattern based on the detection data collected during the predetermined period,
Further comprising the steps of: detecting an event of a disaster by using the LED illumination system;

Images