JP2608512B2 - Fire detection method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a fire detection method for detecting a fire by detecting infrared rays emitted from a fire source. And fire detection methods to clarify. The present invention provides a method for performing effective fire detection regardless of a fire state such as a smoldering fire or a flaming fire.

[0002]

2. Description of the Related Art Conventionally, a flame detecting method for detecting infrared rays emitted from a flame has been put to practical use. The mainstream of these flame detection methods is to detect characteristic spectral lines (4.4 μm band; resonance emission band of CO ₂ ) radiated from the flame. Such attempts have been proposed. For example, Japanese Patent Application Laid-Open
No. 2497 detects a radiation dose at 4.3 μm and two wavelengths before and after 4.3 μm, and determines that the flame is present when the radiation dose at 4.3 μm and the other two wavelengths becomes a certain value or more. Japanese Patent Application Laid-Open No. 57-96492 proposes that the occurrence of a flame is detected by determining whether or not a valley exists between two convex portions.

In addition, Japanese Patent Application Laid-Open No. 61-32195 discloses a first radiation detecting means for detecting radiation having a wavelength in the near infrared region, and a second radiation detecting means for detecting radiation having a wavelength in the photographic infrared region. Receiving means for receiving output signals from the first and second radiation detecting means, and calculating a logical combination of the output signals based on a level difference between these output signals and synchronism; and a combined output signal from the calculating means. Discloses a fire detecting device provided with a detecting means for distinguishing between a fire signal and a noise signal. This is because the flaming fire and the visible light noise have a synchronism in the correlation between the infrared rays of 2.3 μm and 0.9 μm, the smoldering fire does not show the synchronization, and the flaming fire and the smoldering fire are close. The infrared light intensity is higher than the photographic infrared light intensity, and the visible light noise is based on the fact that the near infrared light intensity is smaller than the photographic infrared light intensity. It distinguishes burning fire.

[0004] However, in the method of determining that there is no fire when the radiation intensity in the visible or near-infrared region is higher than the radiation intensity in the infrared region such as electric lamps, false alarms due to ordinary electric lamps are reduced. If the heating element does not emit visible or near-infrared rays or is weak, it is determined to be a fire and a false alarm is issued. That is, an electric heater or the like generates a false alarm, and its application is greatly restricted. Also, 4.3
μm and the radiation dose at two wavelengths before and after
In the method of judging as a flame when the radiation dose at 4.3 μm and the other two wavelengths exceeds a certain value, the flame can be detected but the flame is derived from a fire or a useful heat source. Origin cannot be detected. That is, there is a defect that a false alarm is generated by a flame of a gas range, a gas stove, or the like. Further, in the case of comparing the level difference between two types of radiation of 2.3 μm and 0.9 μm and the synchronicity to distinguish between fire and visible light noise, and to distinguish between flaming fire and smoldering fire, the type of fire However, there is no guarantee that the synchronicity mentioned here is observed depending on the way of burning, and the reliability is lacking.

In response to such a situation, the present inventors have detected infrared rays radiated from an infrared source in a plurality of wavelength bands,
A highly reliable fire detection method is established by providing a signal processing circuit for judging whether or not a fire has occurred based on the detection output of the infrared detector and the temporal change in the ratio of the detection output.

For example, the present inventors have disclosed in Japanese Patent Application No. 03-348.
547 (Japanese Patent Application Laid-Open No. 05-159174), a short-wavelength infrared wavelength band mainly detecting relatively high-temperature heat generation, a long-wavelength infrared wavelength band mainly detecting relatively low-temperature heat generation, and the presence or absence of a flame. Each of these wavelength bands is used to determine the emission spectrum of combustibles in the smoldering state, various losses in the transmission of infrared radiation through the atmosphere, and the use of fire detectors. 2.8 μm to 3.0 μm in consideration of the loss in the housing required for the configuration.
2 μm, 4.2 μm to 4.6 μm, 4.6 μm
The present invention provides a fire detection method in which four wavelength bands of 5.5 μm and 8.0 μm to 10.0 μm are specified and detected. These wavelength bands are two wavelength bands as a wavelength pair for detecting high-temperature heat generation, two wavelength bands as a wavelength pair for detecting low-temperature heat generation, and a wavelength for detecting the presence or absence of a flame. It is a belt.

[0007] Japanese Patent Application No. 03-34847 (Japanese Unexamined Patent Application Publication No.
In 159174), the temperature of the infrared light source is calculated based on the ratio of the infrared light intensity in each of the above-mentioned wavelength bands, the infrared light intensity in any one of the above-mentioned wavelength bands is obtained from this temperature, and this infrared radiation intensity and the wave-length band are detected. The fire condition is determined by calculating the heating area based on the output of the infrared detector as follows.

[Lambda] 1, .lambda.2,... .Lambda.n (n = an integer of 2 or more), and the detection outputs of the respective wavelength bands detected by the infrared detecting section D are V1, V2,. It is assumed that these detection outputs accurately reflect the infrared intensity of each wavelength band incident on the infrared detection unit D. By the way, according to Planck's radiation law, the radiation intensity per unit area of the infrared radiation emitted from an object at a certain temperature T into a half space at a wavelength λ is expressed by the following equation.

[0009]

(Equation 1)

Here, C1 and C2 are C1 = 2πh
c ² and C ² are constants determined by hc / k. Where h
Is the Planck constant, c is the light speed, and k is the Boltzmann constant.

In the above equation (1), two detection wavelength bands λ1,
By substituting λ2 and the radiated infrared intensities P1 and P2 in the wavelength band to derive an approximate expression for obtaining the temperature T,

(Equation 2) Here, assuming that there is no absorption between λ1 and λ2 from the infrared source S to the infrared detecting unit D, P1 and P2 in the above equation (2)
P2 can be replaced by V1 and V2. That is,

(Equation 3) Becomes From the equation (3), the temperature of the infrared source is obtained by detecting the infrared rays having two different wavelengths.

Next, from the temperature T obtained by the above equation (3), the blackbody radiation intensity per unit area at λ1 or λ2 (this is defined as P1 ′ or P2 ′) is represented by Planck's radiation law, ie, (1) ) Formula On the other hand, the intensity of the infrared light incident on the infrared detection unit D becomes ππL ² depending on the distance L from the infrared light source S. Actually, since s is unknown, for convenience, the intensity of infrared light incident on the infrared detector DT from the infrared source is used as the calculated value of P1 ″ and P2 ″. Since it is assumed that the output of the infrared detection unit accurately reflects the intensity of the incident infrared light, P1 or P2 can be obtained from V1 or V2. Therefore, the infrared intensity P1 "or P2" to be incident on the infrared detector D from an infrared source having an area (s times the unit area) having the temperature T obtained above is P
1 'or P2' is a value obtained by multiplying the 2PaiL ² and s to. That is, P1 ″ = P1 ′ × 2πL ² × s (4) P2 ″ = P2 ′ × 2πL ² × s (5) Here, since it is assumed that the output of the infrared detector accurately reflects the intensity of the incident infrared light, V1 or V2
From the above, P1 or P2 can be found using the above equations (3) and (2). Therefore, if the distance L is known, the ratio between the actually detected incident infrared intensity P1 or P2 and the calculated incident infrared intensity P1 ′ or P2 ′ can be found from equations (4) and (5). Represents the area s of the infrared source S.

Furthermore, by using the wavelength band for detecting the resonance radiation band of CO _2, infrared black body radiation in resonance radiation band CO ₂ from the temperature and heating area of the infrared source determined by said means by equation (1) calculating the intensity Pco ₂ ', determined infrared intensity Pco ₂ "to be incident on the infrared detection portion in the same manner as such the P1, calculates a ratio of the actual and observed Pco ₂ thereto. here, Pco ₂ ">> In the case of Pco ₂ , the infrared light source is accompanied by a flame. In this way, the situation of the fire is grasped and a fire alarm is issued.

[0014]

However, in the above fire detection method, there is a slight difference in the detection result between the heating element without flame and the heating element with flame. That is, in a heating element with a flame, the resonance radiation of CO ₂ (4.4 μm band), which is a characteristic radiation of the flame, is very strong, but various other radiations around 5 μm due to the radiation of H ₂ O etc. It has been found to appear. As a result, in the case of a heating element with a flame, there has been a problem that the scale of the heating element cannot be accurately determined.

[0015]

Means for Solving the Problems In order to achieve the above object, according to the present invention, (1) detecting infrared intensity in a plurality of infrared radiation bands including a resonance radiation band of CO ₂ from a detected object, In a fire detector that determines whether or not a fire has occurred based on the relative ratio and absolute value of the detected values and their temporal changes, the relative or absolute value of the value obtained by subtracting the characteristic radiant component of the flame from the intensity of the detected infrared light A fire detection method comprising calculating a temperature and an area of a detection target from a value.
(2) The plurality of infrared radiation bands are from 2.8 μm to 3.2.
μm, 4.2 μm to 4.6 μm, 4.6 μm to 5.
The fire detection method according to the above (1), wherein the diameter is 5 μm, 8.0 μm to 10.0 μm. (3) Infrared intensity of 4.2 μm to 4.6 μm and 4.6 μm
The method for detecting a fire according to the above (2), wherein the temperature and the area of the object to be detected are calculated by using a value obtained by subtracting a characteristic radiation component of the flame from the intensity of the infrared ray of about 5.5 μm. Was invented.

[0016]

DETAILED DESCRIPTION OF THE INVENTION Hereinafter, the present invention will be described specifically. FIG. 1 shows one configuration example of a fire detector to which the present invention is applied. In this example, a chopper periodically cuts off infrared radiation emitted from a fire or similar heat source,
Four different wavelength bands are detected by the four pyroelectric infrared sensors. These sensors have a built-in bandpass filter that transmits a predetermined wavelength band. However, even if there are four or less pyroelectric infrared sensors, a method may be used in which four wavelengths are detected by using the switching means. Further, even if only four or less pyroelectric infrared sensors are used, a system may be used in which four kinds of wavelength bands are detected using the switching means. The signal detected by each sensor is amplified by an amplifier circuit and then converted into a digital signal by an A / D converter. The microprocessor performs synchronous detection and filtering on the digital signal based on the chopper dividing cycle, and returns the signal divided by the chopper to a continuous signal sequence. Further, the signal sequence obtained here is converted into a signal sequence for communication and digitally transmitted to a host computer.

The wavelength band of infrared rays detected in the example of FIG. 1 is a wavelength band for efficiently detecting the radiation of the heating element from a low temperature range to a high temperature range as shown in FIG. That is, the center wavelength 3 μm half width 0.4 μm, the center wavelength 4.4 μm half width 0.4
μm, center wavelength 5.5μm, half width 0.8μm, center wavelength 8.5μ
m The half width is 1.0 μm. In the above-mentioned wavelength band, radiation of the heating element in the high-temperature region is mainly detected, and the heating element in the high-temperature region of 400 ° C. or higher is monitored in combination with the infrared intensity detected in the wavelength band. For the wavelength band, the presence or absence of a flame is monitored. The wavelength band is a wavelength band that efficiently detects from low temperature to high temperature, and in combination with the wavelength band, monitoring of heating elements in high temperature range,
Monitoring of the heating element in the low temperature range is performed in combination with the wavelength band.
In the wavelength band, radiation of a low-temperature heating element of 400 ° C. or less is efficiently detected, and the heating element in a low-temperature area of 400 ° C. or less is monitored in combination with the wavelength band. In the process from the smoldering state to the fire, the low-temperature heating element expands while gradually increasing in temperature. Therefore, it is necessary to monitor the temperature of the heat source in a wide temperature range from a low temperature to a high temperature.

The radiation component peculiar to the flame has a substantially constant wavelength band component irrespective of the size and quality of the flame, but the amount contained in the entire radiation varies depending on the size and quality of the flame. The radiation of the flame is divided into the radiation part of the black body in the flame and the radiation part peculiar to the flame, and is the sum of the radiation part of the black body part and the radiation part peculiar to the flame. Furthermore, the amount of the radiation component peculiar to the flame contained in the entire radiation varies depending on the size of the flame. Here, the wavelength band in which the radiation peculiar to the flame appears is a wavelength band mainly serving as the resonance radiator of CO ₂ , and the radiation peculiar to the flame also appears in the wavelength band. The radiation appearing in the wavelength band is considered to be moisture generated during combustion.

FIG. 3 shows that the temperature of the detected object is at least CO ₂
This is an example showing the relative ratio of the intensities of a plurality of infrared radiation bands including the intensity of infrared radiation in the resonance radiation range of FIG. 1 when an experiment was conducted to detect various heating elements in a fire detector having the configuration shown in FIG. It shows the relative ratio of sensor output in each wavelength band. The radiation component peculiar to the flame has a component ratio of the FR point in FIG. 3 in the wavelength band. With this point as the center, the ratio of sensor outputs due to the combustion of different combustibles is on a straight line. In other words, the emission component of all types of flame in the wavelength band and in the wavelength band is FR
The component ratio is represented by a point. If you subtract the amount that the flame has, it will gather at a point on the blackbody radiation line regardless of the size of the flame.

[0020] The emission of flame refers to the C generated by combustion.
Resonance radiation due to molecular vibrations such as O ₂ and H ₂ O and heat radiation due to solid particles such as graphite generated by combustion are superimposed. The radiation of the solid particles is distributed over the entire flame and is a radiation of the translucent body as it is radiation by particles dispersed in the gas. And even if this part is a translucent body,
If the flame is large enough, the emissivity can be considered to be approximately 1. In addition, radiation peculiar to the flame due to CO ₂ , H ₂ O, etc. has a very high radiation efficiency and is radiation from the surface of the flame. Therefore, for example, radiation accompanying the combustion of normal heptane is drawn between point a, which is heat radiation due to solid particles such as graphite generated by combustion, and point FR, which is radiation specific to a flame due to CO ₂ , H ₂ O, and the like. It has a radiation component on a straight line. Similarly, radiation accompanying the combustion of methanol becomes a radiation component on a straight line between the point b and the FR point.

The temperature of the flame is highest for the gas in the outer layer of the flame. Therefore, the temperature at which heat radiation occurs is often not the maximum temperature of the flame, but when considering the scale of the entire flame, radiation of the entire flame, that is, solid particles such as graphite generated by combustion according to the above considerations It is better to consider heat radiation due to

A calculation method for obtaining this temperature will be discussed below. In the example of FIG. 3, the FR point, which is the origin of a straight line indicating the characteristics of the flames of normal heptane and methanol, has a component ratio of wavelength band: wavelength band = 74: 26. Therefore, it is only necessary to consider a phase diagram having this point as a vertex.

Considering the above, it is possible to separate the heat radiation component in the flame by subtracting the flame-specific radiation represented by the FR point from the actually detected flame radiation component. It will be on the curve of the blackbody radiation inside. By performing such processing, the temperature can be obtained from the ratio of the wavelength bands to the wavelength band by the calculation of Expression (3) in the same manner as the normal calculation of the temperature of blackbody radiation. Therefore, the radiation component of the flame may be subtracted from the wavelength band.

Hereinafter, a specific calculation example will be described. In this calculation example, an approximate calculation is performed in order to simplify the calculation. Examining FIG. 3 and the like, it can be seen that the characteristic of the curve of the blackbody radiation is that the ratio of the wavelength band has little change with respect to the temperature and is about 20% of the total of the three wavelengths. Therefore, by subtracting the component peculiar to the flame so that the ratio of the wavelength band becomes about 20% of the sum of the three wavelengths of the wavelength band-wavelength band-wavelength band, it is possible to leave a substantially black body radiation component. here,
700 ° C to 1000 ° C based on the ratio of the wavelength band after subtraction
The error can be reduced by adjusting to the blackbody radiation of.
In this case, the error given to the temperature calculation result is at most 20.
It is about ° C. If this is adjusted to blackbody radiation of 500 ° C. or less, the error at high temperatures increases.

The calculation procedure will be described below. First, before calculating the relative ratio of the intensities of a plurality of infrared radiation bands including the intensity of infrared radiation in the resonance radiation region of CO ₂ at least from the temperature of the detection target, a radiation component specific to the flame is subtracted. For example, the output values of the sensors for each of the components in three wavelength bands are V
1, V2, and V3, and the value subtracted as a flame-specific component is Vx. Vx is the sum of the wavelength band and the component peculiar to the flame of the wavelength band.
Vf3. At the point c in FIG. 4, Vx corresponds to the length from the point c on the extension of the straight line from the FR point to the point c to the curve of the blackbody radiation, of which the component of the wavelength band is Vf2.
The component in the wavelength band is Vf3. Also, if the ratio of the wavelength band in the blackbody radiation as a reference is R2, Therefore, Becomes The emission component of the flame may be subtracted from the wavelength band and the wavelength band using Vx obtained here. The temperature of the flame is determined by applying the value of the wavelength band and the value of the wavelength band corrected here to the equation (3). The area and the CO ₂ ratio are calculated from the temperature obtained here by applying Equations (4) and (5) to the area for the wavelength band and the CO ₂ ratio for the wavelength band.

With the above-described means, the temperature and area of the infrared source are determined. By the way, even if the temperature of the infrared source is high, it is not a fire if the size, that is, the area is small, and if the temperature is low, the possibility of a fire is high if the area is large. What is important in considering such a heat source in all states is the amount of heat that the heat source gives to the surroundings. That is, if the amount of heat given to the surroundings is large, the risk of fire spread increases, and the expansion speed also increases. The amount of heat the fire source gives to the surroundings is determined by the amount of radiant energy of the fire source. Therefore, the amount of radiant energy can be considered as the scale of the fire source. The radiant energy W of a heat source having a certain temperature T and an area s is expressed as W = T ⁴ σ · s (W) (8) according to Stefan-Boltzmann's law. Here, σ is a Stefan-Boltzmann constant, and σ = 5.673 × 10 ⁻¹² (W / cm ² · deg.
⁴ ). Therefore, the scale of the fire source is determined from the temperature and area of the infrared source.

In this embodiment, an alarm is issued when the amount of radiant energy obtained by the above means exceeds the amount of heat generated in a dangerous state in the target space. It is said that a general fire is in a dangerous state when the calorific value of the fire source is 5 kW to 20 kW, that is, a state in which a person present in the space feels dangerous. Of course, this value has a characteristic that can be changed depending on the size of the target space, the use, and the like.
In addition, one or more subdivisions are provided for values smaller than the heat value of the danger state, and the danger is divided into stages each time the amount of radiated energy exceeds each division, and an alarm is issued. From the state, a warning alarm can be issued, and not only a more accurate fire detection is possible, but also an initial response is facilitated.

Further, by focusing on the increase in the amount of radiant energy, a change in the state of the fire can be grasped more accurately. That is, when the amount of radiant energy is large and the amount of radiant energy is increasing, the possibility of a fire is very high. If the amount of radiant energy is large but there is no change over time, it is not a fire but a useful flame such as heating in many cases. Therefore, by issuing an alarm based on the rate of increase in the amount of radiant energy, it is possible to reliably detect a fire from an earlier stage.

The rate of increase in radiant energy is determined from the rate of change of the calculated amount of radiant energy over a certain period of time. As this method, a method of storing calculated values of radiant energy in a time series for a predetermined time, performing high-frequency cutoff filter processing such as a moving average on the time series values, and comparing two radiant energy values separated by a certain time, Based on the time series values stored for a certain period of time, a method of obtaining a tendency of change from these values by the least square method is used. The storage time when calculating the increase rate is desirably 10 seconds or more, but desirably 3 minutes or less to avoid a delay in the alarm.

The criterion for determining a fire is that when the rate of increase in the amount of radiant energy exceeds a predetermined value, when the rate of increase in the amount of radiant energy is divided into one or more predetermined steps, In some cases, the rate of increase in the amount of radiant energy of the heat source may be divided into one or more predetermined stepwise divisions of the radiant energy of the heat source. These are appropriately selected according to requirements determined by the space or the like. Usually, this division is appropriately about 1 to 5. The warning is determined by the radiant energy amount according to each step of the increase rate when the increase rate of the radiant energy amount exceeds the above-mentioned standard or when the heat source is determined to be a fire when the increase falls within the above category. There are cases where the alarms in the six stages are shifted to higher-level alarms, which are also selected as appropriate according to the specifications of the target space.

[0031]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS As an embodiment, an example in which the fire detection method according to the present invention is applied to a fire detector having the configuration shown in FIG. In this embodiment, the detection wavelength band of each sensor of the detection unit shown in FIG. 1 is 3.0 ± 0.2 μm, 4.4 ± 0.2 μm, 5.0 ± 0.4 μm.
m, 8.5 ± 0.5 μm. Formulas (1) to (8) are stored in the host computer, and the temperature, area, presence / absence of flame, and radiant heat of the heating element are calculated. When the output of the sensor is less than 100 times the noise level, the host computer performs a filtering process for noise reduction. The time constant of the filtering process is 8 seconds when the output of the difference sensor is 100 times or less the noise level, the time constant is increased when the output of the sensor is small, and 64 seconds when the output of the difference sensor is 10 times or less. And Judgment of fire is made in six steps based on the amount of radiated heat. Further, the rate of increase of the radiant heat is obtained by the least square method, and when this is 1 W / sec or more, it is determined that the heating element is expanding,
If it is 10W / sec or more, it is rapidly expanding. When the fire is being expanded, the classification of the fire judgment is advanced to the one-stage dangerous side, and when the fire is rapidly expanding, the fire determination is advanced to the two-stage dangerous side. The storage time when performing the least squares method is 20 seconds when the output value of the sensor is 100 times or more the noise level, but when the output of the sensor is small, the storage time is increased, and the storage time of the sensor is increased. If the output is 10 times or less the noise level, the time is set to 3 minutes. In this embodiment, the FR point in FIG. 3 is set as wavelength band: wavelength band: = 74: 26. Also, R in equation (6)
2 is set based on 700 ° C. blackbody radiation.

FIG. 5 shows a comparison of how the size of a heating element is determined between a conventional fire detection method and the fire detection method of the present invention by experiments. This experiment utilizes normal heptane combustion. Burn normal heptane in square dishes of various sizes and burn 10m
This is a case where the distance is detected from the distance. In the conventional fire detection method, a large error occurs when the scale of the heating element is small. However, it can be seen that the error is small in the present embodiment. If such an error occurs, the scale of combustion in the initial stage of the fire cannot be grasped, causing a delay in the initial response. On the other hand, according to the fire detection method of the present invention, it can be seen that the error is small even when the scale of the heating element is small.

[0033]

According to the fire detecting method of the present invention, the scale of a heating element with a flame can be accurately determined. FIG. 5 is a comparison between the conventional fire detection method and the fire detection method of the present invention, in which the size of the heating element is determined by experiments. This experiment utilizes normal heptane combustion. Normal heptane in various sizes of square dishes was burned and detected from a distance of 10 m. It can be seen that in the conventional fire detection method, a large error occurs when the size of the heating element is small.
If such an error occurs, the scale of combustion in the initial stage of the fire cannot be grasped, causing a delay in the initial response. On the other hand, according to the fire detection method of the present invention, it can be seen that the error is small even when the scale of the heating element is small. Therefore, appropriate measures can be taken in the initial stage of the fire.

[Brief description of the drawings]

FIG. 1 shows an example of a detector according to the invention.

FIG. 2 shows a detection wavelength band according to the present invention.

FIG. 3 shows a relative ratio of sensor output in each wavelength band when an experiment for detecting various heating elements is performed in the fire detector having the configuration shown in FIG.

FIG. 4 is a reference diagram of a processing procedure formula according to the present invention.

FIG. 5 is a comparison between a conventional fire detection method and a fire detection method according to the present invention for determining the combustion scale of normal heptane.

Claims (3)

(57) [Claims] An infrared intensity in a plurality of infrared radiation bands including a resonance radiation band of CO ₂ from an object is detected,
In a fire detector that determines whether a fire has occurred based on the relative ratio and absolute value of the detected values and their temporal changes, the relative or absolute value of the value obtained by subtracting the characteristic radiant component of the flame from the intensity of the detected infrared light A fire detection method comprising calculating a temperature and an area of a detection target from a value. 2. The plurality of infrared radiation bands are 2.8 μm.
３3.2 μm, 4.2 μm to 4.6 μm, 4.6 μm
The fire detection method according to claim 1, wherein the diameter is from m to 5.5 m, and from 8.0 m to 10.0 m. 3. The temperature of the object to be detected using a value obtained by subtracting a characteristic radiant component of a flame from the intensity of the infrared light of 4.2 μm to 4.6 μm and the intensity of the infrared light of 4.6 μm to 5.5 μm. 3. The method according to claim 2, wherein the area is calculated.