KR20210049661A - Fire detection apparatus and method using light spectrum analysis - Google Patents

Abstract

The present invention provides an apparatus and method for detecting smoke, which analyzes spectral distribution of secondary light scattered or transmitted by smoke particles in order to differentiate between smoke from daily life and smoke from a real fire, thereby reducing the generation of an alarm of a class B fire. The apparatus (100) for detecting fire according to the present invention comprises one or a plurality of light-emitting units (110), a light-receiving unit (120) for detecting optical spectrum, and fire determining unit (160). When smoke is introduced into the apparatus (100) for detecting fire in case of the fire, primary light (130) emitted by the light-emitting units (110) is scattered or transmitted by smoke particles (140) and then is incident as secondary light (150) on the light-receiving unit (120). The light-receiving unit (120), which receives secondary light, is capable of detecting optical spectrum, so the light-receiving unit (120) outputs the spectrum (170) of the secondary light (150). The fire determining unit (160) receives and analyzes the spectrum (170) of the secondary light outputted by the light-receiving unit (120) to determine whether the smoke particles introduced inside are smoke particles from smoke from daily life or smoke from fire, thereby distinguishing whether there is fire or not.

Description

Fire detection apparatus and method using light spectrum analysis

The present invention relates to a fire detection technology, and more specifically, to a fire detection device and method for reducing non-fire reports by dividing fire smoke generated by an actual fire and living smoke generated in daily life.

The fire detector is a device that detects fire early by recognizing heat or smoke generated during a fire. When a fire occurs, it is a fire monitoring device that automatically detects the fire and notifies a fire alarm. Fire detectors include heat detectors, smoke detectors, complex detectors, and flame detectors. The heat detector is divided into a differential type that detects fire when the temperature rises rapidly, a constant temperature type that detects fire when the temperature rises above a specified temperature, and a compensation type that uses both differential and constant temperature types. There is a hyperdistributed type. Smoke detectors operate when detecting smoke generated in a fire.There are two types: an ionization type, which uses a change in ion current when smoke enters the detection unit, and a photoelectric type, which uses a change in the amount of incident light of a photoelectric element when smoke enters the detection unit. . The combined hot smoke detector detects heat and smoke at the same time with a compensation heat detection function and a photoelectric smoke detection function. In addition, the flame detector operates when the change of flame in a fire exceeds a certain amount, and operates by the change in the amount of light received by the light-receiving element by a local flame, and there are ultraviolet type, infrared type, ultraviolet infrared ray combination type, and complex type flame detector. .

In order to detect fires in houses, buildings, etc., the fire detector is installed by attaching a base to the ceiling, wall, etc., and assembling a sensing unit composed of a circuit on the base. In the event of a fire, the fire detector detects flame, smoke, temperature, etc. and sends a signal to the outside to trigger an alarm.

1 is a schematic diagram for explaining the principle of a general photoelectric fire detector that uses a change in the amount of incident light of a photoelectric device as smoke enters the detector. The photoelectric fire detector 10 has an infrared light emitting part 11 and a light receiving part 12 of about 900 nm inside, so that the light 13 emitted by the light emitting part 11 is transmitted to the light receiving part 12. It has a structure in which the light-receiving unit 12 reacts when it is incident on. Since the light-receiving part 12 is arranged to deviate from the path of the light 13 irradiated from the light-emitting part 11, the light 13 of the light-emitting part 11 enters the light-receiving part 12 in a daily environment without smoke. It doesn't work.

FIG. 2 is for explaining a process in which a fire detection operation occurs due to smoke entering the photoelectric fire detector 10 of FIG. 1. As mentioned above, since the light-emitting part 11 and the light-receiving part 12 are arranged so as to deviate from each other, the light 13 is not incident on the light-receiving part 12 in a normal environment, but smoke is generated inside the photoelectric fire detector 10. Upon entering, a part of the light 13 irradiated from the light emitting unit 11 is scattered by the smoke particles 14 and the scattered light 15 is incident on the light receiving unit 12. The light receiving unit 12 is designed to simply turn on/off or output a high/low signal depending on whether the scattered light 15 is detected.

However, since the general photoelectric fire detector operates only with the occurrence of scattered light 15 due to smoke particles introduced into the interior of the light emitting unit 11 and the light receiving unit 12, it is not fire smoke, but cigarette smoke, cooking smoke, dust, etc. As it reacts and operates, non-fire alarms (non-fire alarms) are frequently issued.

As described above, it is a non-fire report that judges that the fire detector is not a fire due to daily heat caused by sunlight, activator lamps, warmers, etc., and daily smoke such as cigarette smoke, food cooking smoke, and fine dust In order to solve the problem that is frequently issued, the present invention was devised. Accordingly, an object of the present invention is to reduce non-fire reports by distinguishing between fire smoke generated by actual fire and daily life smoke.

To achieve the above object, a smoke detector that analyzes the spectral distribution of light scattered by smoke particles is used. The fire detection apparatus and method based on light spectrum analysis according to the present invention includes at least one light emitting unit, at least one light receiving unit having a light spectrum detection function, and a determination unit for classifying a fire by analyzing the light spectrum.

Specifically, the problem solving means of the present invention are as follows.

-Irradiation of light (primary light) to smoke particles using at least one light emitting part

-A light-receiving unit that detects a spectral spectrum from the secondary light by receiving the secondary light that is scattered by the smoke particles or transmitted through the smoke particles.

-Establish and utilize a large amount of light spectrum database (DB) to distinguish fire/non-fire by identifying smoke generated by fire and non-fire smoke generated in daily life.

-It is possible to use artificial intelligence learning method when analyzing light spectrum

The concept of the present invention will become more apparent through specific embodiments described later with reference to the drawings.

When using the fire detection technology according to the present invention that distinguishes between fire smoke and living smoke based on light spectrum analysis, it is possible to reduce non-fire reports in which the fire detector operates by determining that it is not a fire as a fire.

1 is a diagram illustrating the principle of a general photoelectric fire detector,
FIG. 2 is a schematic diagram showing a process in which smoke is introduced into the photoelectric fire detector of FIG. 1 to operate the detector;
3 and 4 are schematic configuration diagrams of a fire detection device based on light spectrum analysis of the present invention,
5 is a view for explaining the operation of the invention when living smoke particles enter the fire detection device based on the light spectrum analysis of the present invention;
6 is a view for explaining the operation of the invention when fire smoke particles enter the fire detection device based on the light spectrum analysis of the present invention;
7 is an operation flowchart of a fire detection method based on light spectrum analysis of the present invention.

Advantages and features of the present invention, and a method of achieving them will become apparent with reference to the preferred embodiments described in detail together with the accompanying drawings. However, the present invention is not limited to the embodiments described below and may be implemented in various other forms. The examples are merely provided to completely disclose the present invention and to fully inform a person of ordinary skill in the art to which the present invention belongs, and the present invention is defined by the description of the claims. will be.

In addition, terms used in the present specification are for explaining examples and are not intended to limit the present invention. In this specification, the singular form also includes the plural form unless otherwise specified. In addition, the term'comprise, comprising, etc.' used in the specification refers to the presence of one or more other elements, steps, operations, and/or elements other than the mentioned elements, steps, operations, and/or elements, or It is used in the sense not to exclude additions.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. In the description of the embodiments, if a detailed description of a related known configuration or function may obscure the subject matter of the present invention, a detailed description thereof will be omitted.

3 is a schematic configuration diagram for explaining a fire detection device and method using a light spectrum analysis according to the present invention, and is a schematic diagram of a state in which smoke particles do not enter the fire detection device 100. The fire detection device 100 according to the present invention includes one or a plurality of light emitting units 110, a light spectrum detecting light receiving unit 120, and a fire classification determining unit 160. Since smoke particles have not entered the device 100, the primary light irradiated from the light emitting unit 110 is detected by the light spectrum detection and light receiving unit 120 as it is. Therefore, the fire classification determination unit 160 receives the same spectrum as the spectrum of the primary light irradiated by the light emitting unit 110 and analyzes it to determine that the current situation is non-fire.

4 is for explaining the operation of the present invention when a fire occurs and smoke particles are introduced into the apparatus 100.

When smoke enters the fire detection device 100, the primary light 130 irradiated from the light emitting unit 110 is scattered by the smoke particles 140 or passes through the smoke particles 140. The light spectrum detection light receiving unit 120 receives the scattered light or transmitted light ('secondary light') 150. The light-receiving unit 120 outputs the spectrum 170 of the received secondary light 150 because it has a light spectrum detection capability. The fire classification determination unit 160 receives and analyzes the secondary light spectrum 170 output from the light receiving unit 120, and determines whether the smoke particles introduced into the device 100 are particles of living smoke or particles of fire smoke. To distinguish whether there is a fire or non-fire.

Again, the principle of the present invention utilizes that the wavelength spectrum of secondary light is different for each smoke when living smoke and actual fire smoke enter. For example, for each type of smoke, a wavelength size of secondary light scattered or transmitted from smoke particles may be reduced or a wavelength shift may occur. It is the principle of the present invention to analyze this to distinguish between fire and non-fire.

5 is a light spectrum detection and receiving unit 120 of secondary light 150a scattered or transmitted by the life smoke particles 140a when the life smoke particles 140a enter the fire detection device 100 shown in FIG. 4 It is to explain the classification of non-fires by using the smoke classification algorithm 180 built in the fire classification determination unit 160 from the spectrum 170a received and output to.

In addition, FIG. 6 is a light spectrum detection and receiving unit of secondary light 150b scattered or transmitted by the fire smoke particles 140b when the actual fire smoke particles 140b enter the fire detection device 100 shown in FIG. 4. It is for explaining that the fire is classified by using the smoke classification algorithm 180 built in the fire classification determination unit 160 from the spectrum 170b received and output by 120.

5 and 6, the light spectrum of the secondary light 150a when the living smoke particles 140a enters and the secondary light 150b when the fire smoke particles 140b enters is different. It is the fire detection apparatus and method of the present invention. As noted, the spectrum of light consists of approximately 400 nm or less ultraviolet light, 400-700 nm visible light, and 700 nm or more infrared light depending on the wavelength. Here, visible light can be seen by the human eye, but ultraviolet and infrared rays are hardly visible to the human eye.

3 to 5, the light-emitting unit 110 may be configured to generate primary light with one light-emitting device, but may be configured to generate light having a desired plurality of wavelength bands by using a plurality of light-emitting devices. In the latter case, when a plurality of light-emitting devices are used, all light-emitting devices may be continuously and simultaneously driven, each light-emitting device may be sequentially driven, or pulse-driven at the same or random time.

Part of the primary light 130 irradiated from the light emitting unit 110 is scattered by the smoke particles 140a or 140b or the secondary light 150a or 150b transmitted through the smoke particles is detected by the light spectrum detection unit 120 ), the light-receiving unit 120 outputs spectrums 170a or 170b of patterns having different amplitudes for each wavelength band.

The light spectrum detection and light receiving unit 120 may be implemented with a spectrometer. One light spectrum detection light-receiving unit 120 may be used, or a plurality of light spectrum detection light receiving units 120 may be used. In the case of using a plurality of light-receiving elements, a plurality of light-receiving elements for detecting different wavelength bands can be used to detect the spectrum of a specific band desired, or a plurality of light-receiving units for measuring the same wavelength, but the difference between secondary light by different light receiving positions You can also make it detectable. In addition, when a plurality of light spectrum detection light-receiving elements are used, all light-spectrum detection light-receiving elements may be continuously and simultaneously driven, or pulse-driven sequentially for each light-receiving element or at the same or random time.

Next, the fire classification determination unit 160 analyzes the spectrum for each wavelength band of the light (secondary light) detected by the light spectrum detection and light receiving unit 120 using the smoke classification algorithm 180, and determines whether the fire smoke is correct from this. It is distinguished from life acting.

In order for the smoke classification algorithm 180 of the fire classification determination unit 160 to distinguish between fire smoke and living smoke, a database constructed with secondary light spectrum data for various smokes previously irradiated may be referred to. Secondary light spectrum data of various fire smoke particles can be obtained depending on the cause or aspect of the fire, and similarly, various secondary light spectrum data can be obtained for life smoke. By collecting and analyzing these data in advance, it is possible to build a secondary light spectrum database for smoke particles. For reference to the spectrum database, an index such as a light intensity peak value for each wavelength or a distribution position and number of light intensity peak values for each wavelength may be used.

When the smoke classification algorithm 180 is executed, an artificial intelligence learning technology such as a deep neural network can be used. In this case, a learning model is built through machine learning such as deep learning using various secondary light spectra of fire smoke and life smoke as learning data, and whether the light spectrum currently detected from this learning model is from fire smoke or life smoke. You can inference.

7 is a flowchart illustrating an operation of a fire detection method based on light spectrum analysis according to the present invention.

210: A light source (for example, the light emitting unit 110) emits light to irradiate primary light.

220, 230: When smoke is introduced (for example, into the fire detection device 100 of FIG. 3), the primary light is scattered due to the smoke particles or a light spectrum is generated by the transmitted secondary light. The generated light spectrum is detected as a distribution for each wavelength band of a specific pattern (for example, by the light receiving unit 120).

240, 260: Compare the detected spectra with the wavelength band distribution spectrum according to the type of smoke stored in, for example, a secondary light spectrum DB (smoke DB) 250 for smoke described above. Machine learning can be used at this time. A fire alarm is issued by classifying whether the incoming smoke is fire smoke or living smoke through spectrum comparison to determine whether it is a fire or not.

The functions or processes of the components of the present invention described above are among the digital signal processor (DSP), processor, controller, application-specific IC (ASIC), programmable logic device (FPGA, etc.), and other electronic devices. It can be implemented as a hardware element including at least one and a combination thereof. It can also be implemented as software in combination with hardware elements or independently, which software can be stored on a recording medium.

Until now, the present invention has been described in detail through a preferred embodiment of the present invention, but those of ordinary skill in the art to which the present invention pertains to the present invention are different from the contents disclosed in the present specification without changing the technical spirit or essential features. It will be appreciated that it may be implemented in other specific forms. It should be understood that the embodiments described above are illustrative in all respects and not limiting. In addition, the scope of protection of the present invention is determined by the claims described later rather than the detailed description, and all changes or modified forms derived from the scope of the claims and their equivalent concepts should be interpreted as being included in the technical scope of the present invention. do.

Claims (19)

A light emitting unit that irradiates light;
A light-receiving unit configured to detect a light spectrum of a pattern having a different amplitude for each wavelength band of the received secondary light by receiving secondary light scattered or transmitted by the smoke particles of the light irradiated from the light-emitting unit; And
A fire detection device using light spectrum analysis, comprising a fire classification determination unit that determines whether the smoke particle is a particle of living smoke or a particle of fire smoke by analyzing the light spectrum output from the light receiving unit. The fire detection apparatus according to claim 1, wherein the wavelength band of light irradiated from the light emitting unit includes ultraviolet, visible, and infrared bands. The fire detection apparatus according to claim 1, wherein the wavelength band of light irradiated from the light emitting unit includes at least one of ultraviolet, visible, and infrared bands. The method of claim 1, wherein the light emitting unit comprises two or more light emitting devices,
Fire detection device using light spectrum analysis, in which the two or more light emitting devices are simultaneously driven. The method of claim 1, wherein the light emitting unit comprises two or more light emitting devices,
A fire detection device using a light spectrum analysis in which each of the two or more light emitting devices is pulse driven. The fire detection apparatus according to claim 1, wherein the light receiving unit is a spectrometer. The method of claim 1, wherein the light receiving unit
Fire detection device using light spectrum analysis, including two or more light-receiving elements for detecting different wavelength bands. The fire detection apparatus according to claim 7, wherein the two or more light-receiving elements are driven simultaneously. The fire detection device according to claim 7, wherein each of the two or more light-receiving elements is pulse-driven. The fire detection apparatus according to claim 1, wherein the light receiving unit includes two or more light-receiving elements measuring the same wavelength, and is capable of detecting a difference between the received secondary light by different light-receiving positions. The fire detection apparatus according to claim 10, wherein the two or more light-receiving elements are driven simultaneously. The fire detection apparatus according to claim 10, wherein each of the two or more light receiving elements is pulse driven. The method of claim 1, wherein the fire classification determination unit
Fire detection device using light spectrum analysis, referring to a database constructed with various secondary light spectrum data of fire smoke and life smoke in order to determine fire smoke and life smoke. The method of claim 1, wherein the fire classification determination unit
In order to determine fire smoke and life smoke, using a machine-learned learning model using various secondary light spectra of fire smoke and life smoke as learning data, whether the light spectrum detected by the light receiving unit is fire smoke or life smoke Fire detection device using light spectrum analysis to infer whether it is 1) irradiating the smoke particles with light;
2) the irradiated light receives secondary light scattered or transmitted by smoke particles, and detects a light spectrum of a pattern having a different amplitude for each wavelength band of the received secondary light;
3) Analyzing the detected light spectrum to determine whether the smoke particle is a particle of living smoke or a particle of fire smoke, and distinguishing between fire and non-fire, a fire detection method using light spectrum analysis. The method of claim 15, wherein the wavelength band of the light irradiated in 1) includes ultraviolet, visible, and infrared bands. The method of claim 15, wherein the wavelength band of the light irradiated in 1) includes at least one of ultraviolet, visible, and infrared bands. The fire detection method according to claim 15, wherein in 3), in order to determine fire smoke and life smoke, a database constructed from various secondary light spectrum data of fire smoke and life smoke is referred to. The method of claim 15, wherein in order to determine fire smoke and life smoke in 3), detection in 2) is performed using a machine-learned learning model using various secondary light spectrums of fire smoke and life smoke as learning data. Fire detection method using light spectrum analysis to infer whether the generated light spectrum is from fire smoke or life smoke.

Images