JP6682147B2 - Flame detector - Google Patents

Description

  The present invention relates to flame detectors.

  Conventionally, as a radiation type fire detector, a constant radiation type that detects that the radiant energy of a specific wavelength band emitted from a flame has reached a certain amount, a flicker type that detects a flicker peculiar to a flame, and more than one There are various methods such as a two-wavelength method and a three-wavelength method for comparing the magnitude of radiant energy in a wavelength band. Some of these radiant fire detectors detect radiant light such as ultraviolet rays and infrared rays emitted from the flame by a light receiving sensor (for example, a photodiode, a pyroelectric sensor, a discharge tube, etc.).

  As a specific structure of the infrared flame detector, a sensor that captures the carbon dioxide resonance radiation band (near 4.3 to 4.5 μm) among multiple sensors and another sensor on the long wavelength side and / or the short wavelength side is used. One or a plurality of them are provided, and these signals are amplified, AD-converted, and judged by an arithmetic processing device incorporating a flame detection algorithm.

  It is known that a sensor that outputs a signal in accordance with the amount of change in the received infrared light extracts a frequency component (flicker) peculiar to the flame from the frequency component of the background signal and uses it for fire determination.

  On the other hand, if a signal is output according to the absolute amount of received infrared light, the signal from the sensor includes those due to environmental changes (for example, infrared radiation in the background, difference in temperature between morning and evening, etc.), The difference (amount of change) between the environmental noise (a signal that changes relatively slowly) and the signal that changes abruptly is used for flame determination.

  By treating the amount of change of each sensor as a signal, it is possible to perform flame detection independent of the background environment. In addition, basically, when only the signal change amount of the carbon dioxide resonance radiation band is large, the flame is determined.

Japanese Patent Laid-Open No. 1-74696 JP-A-3-263197 JP-A-8-115480 JP, 10-326391, A

  However, in the case of an algorithm that determines a flame when only the amount of signal change in the carbon dioxide resonance radiation band is large, if the flame is erroneously detected even when emitted from hot air containing carbon dioxide or combustion gas such as exhaust gas. There is.

  The present invention has been made to solve the above problems, and an object of the present invention is to provide a flame detector capable of accurately detecting a flame.

  In order to achieve the above object, a flame detector according to the present invention includes a first band filter that transmits infrared light of a carbon dioxide resonance radiation band emitted by a flame and a band having a wavelength shorter than the carbon dioxide resonance radiation band. A second bandpass filter that transmits infrared light in a band including the band and has a band center at a position apart from the band center of the carbon dioxide resonance radiation band, and a band having a wavelength longer than the carbon dioxide resonance radiation band Infrared light transmitted through the first band-pass filter and the third band-pass filter provided at a position where the band center is distant from the band center of the carbon dioxide resonance radiation band are detected while transmitting the infrared light in the band. A first detection element for converting into an electric signal, a second detection element for detecting infrared light transmitted through the second band filter and converting it into an electric signal, and an infrared light transmitted through the third band filter. To detect A third detection element for converting into an air signal, a value of an electric signal detected by the first detection element, a value of an electric signal detected by the second detection element, and an electric value detected by the third detection element A determination unit that determines whether or not flame is detected by comparing signal values, the determination unit including a value of the electrical signal detected by the second detection element and the third detection element. The calculated value corresponding to the value of the electric signal detected by the first detection element, which is obtained from the approximate straight line obtained from the value of the electric signal detected by, is calculated as the noise amount, and the calculated noise amount is set to the calculated noise amount. Accordingly, the detection sensitivity is changed to determine whether or not the flame is detected.

  According to the flame detector of the present invention, the first detection element detects the infrared light that has passed through the first band-pass filter that transmits infrared light in the carbon dioxide resonance radiation band and converts the infrared light into an electric signal. The second detection element detects infrared light that has passed through the second bandpass filter that transmits infrared light in a wavelength band shorter than the carbon dioxide resonance radiation band, and converts the infrared light into an electric signal. The third detection element detects infrared light that has passed through the third bandpass filter that transmits infrared light in a wavelength band longer than the carbon dioxide resonance radiation band, and converts the infrared light into an electric signal.

  Then, the determination unit compares the value of the electrical signal detected by the first detection element, the value of the electrical signal detected by the second detection element, and the value of the electrical signal detected by the third detection element. Then, it is determined whether or not the flame is detected. At this time, the electricity detected by the first detection element, which is obtained from the approximate straight line obtained from the value of the electric signal detected by the second detection element and the value of the electric signal detected by the third detection element, The calculated value corresponding to the value of the signal is calculated as the noise amount, and the detection sensitivity is changed according to the calculated noise amount to determine whether or not the flame is detected.

  According to the flame detector of the present invention, the infrared light in the carbon dioxide resonance radiation band, the infrared light in the wavelength band shorter than the carbon dioxide resonance radiation band, and the infrared light in the wavelength band longer than the carbon dioxide resonance radiation band. By detecting each of them, converting them into an electric signal, and comparing the values of the respective electric signals to determine whether or not the flame is detected, it is possible to detect the flame with high accuracy. To be

It is a block diagram which shows the structure of the flame detector which concerns on the 1st Embodiment of this invention. It is a block diagram which shows the structure of the 1st arithmetic processing part and 2nd arithmetic processing part of the flame detector which concerns on the 1st Embodiment of this invention. It is a figure for demonstrating the method of setting a correction coefficient. (A) Electric signal of each sensor, (B) Electric signal of each sensor corrected for variations, (C) Electric signal of each sensor corrected for individual variation , And (D) are offset-corrected electrical signals of each sensor. (A) It is a figure which shows a monitoring environment signal value and a live value, and (B) is a figure which shows the amount of change. It is a figure which shows distribution of the amount of change in heat radiation, a flame, and a hot air. (A) It is a figure which shows the amount of change of each zone, and (B) is a figure for demonstrating the method of determining whether the flame was detected. It is a figure which shows a threshold value table. It is a figure for demonstrating the method of performing a fire determination. It is a flowchart which shows the fire determination processing routine in the 1st arithmetic processing part and the 2nd arithmetic processing part of the flame detector which concerns on the 1st Embodiment of this invention. It is a flowchart which shows the fire determination processing routine in the 1st arithmetic processing part and the 2nd arithmetic processing part of the flame detector which concerns on the 1st Embodiment of this invention. It is a figure for explaining a method of changing sensitivity. It is a block diagram which shows the structure of the 1st arithmetic processing part and 2nd arithmetic processing part of the flame detector which concerns on the 2nd Embodiment of this invention. It is a graph which shows the radiant energy according to the wavelength of a black body. It is a figure which shows a threshold value table. It is a figure which shows the example which changes a threshold value according to inclination. It is a figure which shows the example which changes a threshold value according to the strength of an electric signal. It is a flowchart which shows the fire determination processing routine in the 1st arithmetic processing part and 2nd arithmetic processing part of the flame detector which concerns on the 2nd Embodiment of this invention. It is a flowchart which shows the fire determination processing routine in the 1st arithmetic processing part and 2nd arithmetic processing part of the flame detector which concerns on the 2nd Embodiment of this invention.

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

<Outline of Embodiment of the Present Invention>
According to the conventional method using a pyroelectric sensor (for example, Patent Documents 1 and 2), signal processing such as noise cutting and frequency decomposition is required, and there are problems in speeding up the processing and delaying detection time.

  The flame detector of the present embodiment uses a thermopile sensor and processes an electric signal of a direct current component according to the strength of an infrared signal, so that frequency decomposition processing is unnecessary, and high sensitivity and early detection are possible. It is possible.

  However, in the conventional method using the thermopile sensor (for example, Patent Document 4), there is a case where malfunction occurs due to high temperature hot air. Therefore, the number of infrared wavelengths to be detected is provided at least three, and the signals of two (or more) sensors other than the carbon dioxide resonance radiation band are compared to determine whether or not hot air is present. .

  Specifically, the signals output from three infrared sensors equipped with bandpass filters (bandpass filters) that transmit only the bands near 4.0 μm, 4.5 μm, and 5.0 μm are amplified, and the signals are subjected to AD conversion and then processed by the CPU. Input the signal to.

  A flame detection algorithm is incorporated in the CPU, and the flame detection algorithm performs flame determination based on the characteristics of signal changes output from the three infrared sensors. Further, an algorithm for discriminating between hot air and flame is also incorporated in the CPU.

  It is known that when organic matter burns with a flame, it emits infrared rays in the resonance radiation band (4.5 μm) of carbon dioxide. Therefore, by comparing the signal intensities of the band near 4.5 μm, the other bands near 4.0 μm and the bands near 5.0 μm, it is possible to determine whether or not the flame is detected.

  However, even in the case of high-temperature airflow containing carbon dioxide, radiation in the band near 4.5 μm was emitted, and in the flame detection algorithm for high-sensitivity and early detection, there was a case where malfunction occurred due to high-temperature hot air.

  Therefore, in the present embodiment, signals in the band near 4.0 μm and in the band near 5.0 μm are also compared, and the signal in the band near 4.0 μm is larger than the signal in the band near 5.0 μm (with high-temperature radiation). It was decided to provide a means for performing flame judgment in at least two cases, that is, the case and the other case.

  That is, if the signal change amount in the band near 4.0 μm ≦ 5.0 μm, the fire judgment is made strictly.

<First Embodiment>
<System configuration>
The flame detector according to the first embodiment of the present invention will be described below.

  As shown in FIG. 1, the flame detector 10 according to the first embodiment includes a first sensor 12 for detecting infrared light in a band near 4.5 μm of a carbon dioxide resonance radiation band generated by a flame, and a carbon dioxide gas. A second sensor 14 for detecting infrared light in a band near 4.0 μm in a wavelength band shorter than the resonance radiation band, and an infrared light in a band near 5.0 μm in a band wavelength longer than the carbon dioxide resonance radiation band. A third sensor 16, an amplification unit 18 that amplifies the signal from the first sensor 12, an amplification unit 20 that amplifies the signal from the second sensor 14, and an amplification unit 22 that amplifies the signal from the third sensor 16. An amplifier 24 for amplifying the signals from the amplifiers 18, 20, 22; an AD converter 26 for converting the signal from the amplifier 24 into a digital value; and a preprocessing and external output unit 32 for flame detection. The first arithmetic processing unit 28 for controlling and the processing for detecting a flame are performed. And second arithmetic processing unit 30, and an external output unit 32.

  The first sensor 12 detects a filter 12A that transmits infrared light in a band near 4.5 μm of the carbon dioxide resonance radiation band generated by the flame and infrared light that has passed through the filter 12A and converts the infrared light into a DC component electric signal. And a detection element 12B that operates.

  The second sensor 14 includes a filter 14A that transmits infrared light in a band near 4.0 μm, which is a wavelength band shorter than the carbon dioxide resonance radiation band, and an infrared signal that has passed through the filter 14A to detect a DC component electric signal. And a detection element 14B for converting into.

  The third sensor 16 detects the infrared light transmitted through the filter 16A, which transmits infrared light in the band near 5.0 μm, which is a wavelength band longer than the carbon dioxide resonance radiation band, and detects the infrared light transmitted through the filter 16A, and detects the DC component electric signal. And a detection element 16B for converting into.

  It should be noted that the same sensor as the first sensor 12 may be further provided in order to reliably capture a weak electric signal for detecting infrared light in the band near 4.5 μm of the carbon dioxide resonance radiation band.

  The detection elements 12B, 14B, 16B are composed of thermopiles, but may be composed of other photovoltaic type elements such as InAsSb elements.

  The amplification units 18, 20, and 22 independently amplify the electric signals of the first sensor 12, the second sensor 14, and the third sensor 16, respectively.

  The amplifying unit 24 includes a switch unit (not shown) that sequentially switches the electric signals individually amplified by the amplifying units 18, 20, and 22 into a single electric signal by switching the electric signals at a fixed time. The electric signals collected into one are selectively amplified according to the strength of the electric signals (high gain when the signal is small, low gain when the signal is large).

  The first arithmetic processing unit 28 and the second arithmetic processing unit 30 are each configured by a CPU, and the functions obtained by dividing the first arithmetic processing unit 28 and the second arithmetic processing unit 30 for each function realizing unit. Explaining in blocks, as shown in FIG. 2, the first arithmetic processing unit 28 includes a signal acquisition unit 40, a correction coefficient setting unit 42, a correction unit 44, and an alarm display unit 46A. In addition, the second arithmetic processing unit 30 includes an average calculation unit 50, a change amount calculation unit 52, a fire determination unit 54, a number of times determination unit 56, an alarm control unit 58, and an alarm output unit 46B.

  The signal acquisition unit 40 calculates the value of the electric signal from the first sensor 12, the value of the electric signal from the second sensor 14, and the value of the electric signal from the third sensor 16 from the signal output from the AD conversion unit 26. get.

  As shown in FIG. 3, the correction coefficient setting unit 42 irradiates the first sensor 12, the second sensor 14, and the third sensor 16 with reference light of infrared light from a reference light source such as a blackbody furnace. Sometimes, based on each value (refer to FIG. 4A) of the electric signal acquired by the signal acquisition unit 40, the correction coefficient is set in advance by the following first to third correction steps. To do.

  In the first correction step, as shown in FIG. 4B, a correction coefficient for equalizing variations in sensitivity among the sensors is set. For example, the difference between the maximum and minimum values of the electric signal from the first sensor 12, the difference between the maximum and minimum values of the electric signal from the second sensor 14, and the electric signal from the third sensor 16. The correction coefficient for processing is set so that the difference between the maximum value and the minimum value of the values of is the same.

  In the second correction step, the correction coefficient set in the first correction step is used, and as shown in FIG. 4C, in order to equalize the variations in sensitivity among the individual flame detectors 10. Set the correction coefficient of. For example, the difference between the maximum and minimum values of the electric signal from the first sensor 12, the difference between the maximum and minimum values of the electric signal from the second sensor 14, and the electric signal from the third sensor 16. A correction coefficient for processing is set so that each of the difference between the maximum value and the minimum value of the value of is the same as the respective reference value.

  In the third correction step, using the correction coefficient set in the first correction step and the second correction step, as shown in FIG. And a correction coefficient for performing offset correction is set for the value of the electric signal from each of the third sensors 16.

  The correction unit 44 compares the value of the electric signal from the first sensor 12, the value of the electric signal from the second sensor 14, and the value of the electric signal from the third sensor 16 acquired by the signal acquisition unit 40. The correction coefficient is set using the correction coefficient set by the correction coefficient setting unit 42, and is output to the second arithmetic processing unit 30.

  The average calculation unit 50 uses the moving average value (for example, the average value of the past 100 seconds) of the values of the electric signals from the first sensor 12 corrected by the correction unit 44 as the monitoring environment signal value of the signal of the first sensor 12. Calculate (see FIG. 5A). Similarly, the average calculation unit 50 calculates the moving average value of the values of the electric signal from the second sensor 14 corrected by the correction unit 44 as the monitoring environment signal value of the signal of the second sensor 14. Similarly, the average calculation unit 50 calculates the moving average value of the values of the electrical signals from the third sensor 16 corrected by the correction unit 44 as the monitoring environment signal value of the signal of the third sensor 16.

  The change amount calculation unit 52 calculates the difference between the live value of the electric signal from the first sensor 12 corrected by the correction unit 44 and the monitoring environment signal value of the signal of the first sensor 12 calculated by the average calculation unit 50. , As the first change amount (see FIG. 5B). In addition, the change amount calculation unit 52 similarly monitors the live value of the electric signal from the second sensor 14 corrected by the correction unit 44 and the monitoring environment signal of the signal of the second sensor 14 calculated by the average calculation unit 50. The difference from the value is calculated as the second change amount. In addition, the change amount calculation unit 52 similarly monitors the live environment value of the electric signal from the third sensor 16 corrected by the correction unit 44 and the monitoring environment signal of the signal of the third sensor 16 calculated by the average calculation unit 50. The difference from the value is calculated as the third change amount.

  Here, the principle of detecting a flame will be described.

  First, as shown in FIG. 6, a signal detected by infrared light in the band near 4.5 μm of the carbon dioxide resonance radiation band emitted by the flame, a signal detected by infrared light in the band near 4.0 μm, and a signal near 5.0 μm Comparing the signals obtained by detecting the infrared light in the band, in the case of thermal radiation, the slope of the spectrum distribution line changes depending on the temperature due to the Wien's displacement law. At 370 ° C, the band around 4.5 μm has a gentle peak, and when the temperature is higher than 370 ° C, it rises to the left, and when it is lower than 370 ° C, it rises to the right.

  In the case of a flame, due to carbon dioxide resonance radiation, the waveform has a peak in the band near 4.5 μm. Further, radiation on the high temperature side (upward to the left) is compounded by high temperature products (such as smoke).

  In the case of hot air, the waveform has a peak in the band near 4.5 μm, but since it does not contain high-temperature products (such as smoke), radiation on the low-temperature side (upward to the right) is compounded.

  FIG. 7 shows a schematic diagram for deriving the determination ratio in the fire determination according to the present embodiment. 8 shows an example of a threshold table used for fire determination.

  As an example, if infrared radiation of a flame is received and the amount of change in signal as shown in FIG. 7A is obtained in each band, fire determination is performed in the following steps (1) to (5).

(1) Derive a linear function expression using the amount of change in the three bands.

(2) A calculated value in each band is calculated using the derived linear function formula. (See the star plot in Figure 7B)

(3) The ratio between the calculated value and the amount of change in each band is calculated as the determination ratio.

(4) Calculate the noise amount. The amount of noise is a parameter that considers the presence of a high temperature body or some infrared radiator.

  In the present embodiment, a calculated value corresponding to a band near the wavelength of 4.5 μm ignoring carbon dioxide resonance radiation is calculated from the approximation line using the amount of change in the band near 4.0 μm and the amount of change in the band near wavelength 5.0 μm. By calculating the quantity, the magnitude of the influence of infrared radiators other than fire is expressed.

  By treating the calculated noise amount as a reference value in the threshold table, it is possible to make a fire judgment that corresponds to changes in the environment where it is difficult to detect a fire or changes in the monitoring target, thus making product specifications more reliable.

(5) Three thresholds for fire determination are selected using the obtained noise amount as a reference value, and the determination values are applied to perform fire determination.

  In the band near 4.0 μm and the band near 5.0 μm, the judgment ratio is below the threshold, and in the band near 4.5 μm, the flame is detected when the condition that the judgment ratio is above the threshold is satisfied. to decide. In the threshold table, the approximation obtained from the three points of the value of the first variation (near 4.5 μm), the value of the second variation (near 4.0 μm), and the value of the third variation (near 5.0 μm) Of the three judgment ratios for the calculated value of the first variation (near 4.5 μm), the calculated value of the second variation (near 4.0 μm), and the calculated value of the third variation (near 5.0 μm) obtained from the straight line A threshold is set.

  According to the principle described above, in the present embodiment, the fire determination unit 54 sets the value of the first change amount (near 4.5 μm) calculated by the change amount calculation unit 52 and the second change amount (near 4.0 μm). Value, and at least one of the values of the third variation (near 5.0 μm) is equal to or greater than the threshold value E, and the calculated value for each band obtained from the approximate straight line of the values of the three variations, and the first variation, It is determined that the flame is detected when the result of comparing the respective ratios of the second change amount and the third change amount with the respective threshold values satisfies a predetermined condition. The threshold value E is the minimum signal change amount of whether or not to perform the flame determination, and is an example of a common threshold value. On the other hand, in the threshold table for flame determination, the threshold values of the respective ratios are examples of the first threshold value, the second threshold value, and the third threshold value.

  Here, when the value of the second amount of change is larger than the value of the third amount of change, the fire determination unit 54 uses the calculated value of the first amount of change as the amount of noise and sets the threshold value for the ratio of each band to the threshold value table. And make a decision. On the other hand, when the value of the second amount of change is less than or equal to the value of the third amount of change, the fire determination unit 54 sets a threshold value (for example, when the noise amount is 0 when the amount of noise is 0) for a predetermined ratio of each band for strict determination. The determination is performed using a threshold value corresponding to the case).

  The determination by the fire determination unit 54 described above is repeatedly executed at regular intervals.

  As shown in FIG. 9, the number-of-times determination unit 56 determines whether the number of times the fire determination unit 54 has continuously detected a flame is equal to or greater than a predetermined number of consecutive times, or within a certain time by the fire determination unit 54. When the number of times that it is determined that the flame has been detected is equal to or greater than a predetermined cumulative number, the monitoring environment signal value (moving average value) is set to a fixed value and a fire signal is output.

  When the fire signal is output from the number-of-times determination unit 56, the alarm control unit 58 controls the alarm display unit 46A and the alarm output unit 46B to notify the fire. For example, the alarm display unit 46A turns on the red LED, the alarm output unit 46B brings the photocoupler into a conductive state, and activates the contact output forming the external output unit 32.

<Operation of flame detector>
Next, the operation of the flame detector 10 according to the first embodiment will be described.

  First, a correction coefficient is set in advance for the flame detector 10 before installation. Specifically, when the reference light source such as the black body furnace irradiates the first sensor 12, the second sensor 14, and the third sensor 16 with the reference light of infrared light, the flame detector 10 The correction coefficient setting unit 42 sets the correction coefficient by the above-described first to third correction steps.

  The flame detector 10 for which the correction coefficient is set is installed at a place where a fire judgment should be made, and an electric signal is output from each of the first sensor 12, the second sensor 14, and the third sensor 16 of the flame detector 10. When the value of each signal is input to the first arithmetic processing unit 28 via the amplification units 18, 20, 22, 24 and the AD conversion unit 26, the first arithmetic processing unit of the flame detector 10 28 and the second arithmetic processing unit 30 repeatedly execute the fire determination processing routine shown in FIGS. 10 and 11 at regular intervals.

  In step 100, the signal acquisition unit 40 determines the value of the electric signal from the first sensor 12, the value of the electric signal from the second sensor 14, the electric value from the third sensor 16 from the signal output from the AD conversion unit 26. Get the value of a signal.

  In the next step 102, the correction unit 44 sets the value of the electric signal from the first sensor 12, the value of the electric signal from the second sensor 14, and the value of the electric signal from the third sensor 16 acquired in step 100. On the other hand, the correction is performed using the correction coefficient set in advance.

  Then, in step 104, the average calculation unit 50, for each of the value of the electrical signal from the first sensor 12, the value of the electrical signal from the second sensor 14, the value of the electrical signal from the third sensor 16, A moving average value is calculated based on the sensor value corrected in step 102 and the sensor value previously corrected in step 102.

  In step 106, the change amount calculation unit 52 performs the above operation on each of the value of the electric signal from the first sensor 12, the value of the electric signal from the second sensor 14, and the value of the electric signal from the third sensor 16. The first change amount, the second change amount, and the third change amount are calculated based on the sensor value corrected in step 102 and the moving average value calculated in step 104.

  Then, in step 108, the fire determination unit 54 determines the first change amount, the second change amount, and the third change amount calculated in step 106, the first change amount calculated in step 106 in the past, and the first change amount. Whether the value of any one of the first change amount, the second change amount, and the third change amount has remained below the negative reference value for a certain period of time based on the second change amount and the third change amount. Depending on whether or not the moving average value is reset, it is determined. When resetting the moving average value, the process proceeds to step 110, and the sensor value corrected in step 102 is used to determine the value of the electric signal from the first sensor 12 and the value of the electric signal from the second sensor 14. , The moving average value is reset for each value of the electrical signal from the third sensor 16.

  On the other hand, if the moving average value is not reset, the process proceeds to step 112.

In step 112, the fire determination unit 54 determines whether or not all of the first change amount, the second change amount, and the third change amount calculated in step 106 are less than a predetermined threshold value E. . When it is determined that all of the first change amount, the second change amount, and the third change amount are less than the predetermined threshold value E, in step 114, if the current mode is the fire mode or the caution output mode. , Shifts to the normal mode and ends the fire determination processing routine. If the current mode is the normal mode, the normal mode is continued.
Note that, as shown in FIG. 12, when an environment change without a flame occurs and the live value deviates from the bias value, the sensitivity is switched to a low sensitivity in order to prevent malfunction due to the noise amount of the environment change. . The sensitivity is changed by changing the minimum detection sensitivity (threshold value E).

  When it is determined that at least one of the first change amount, the second change amount, and the third change amount is equal to or greater than the predetermined threshold value E, in step 116, the fire determination unit 54 causes the second change amount. Is less than or equal to the third change amount. If the second change amount is less than or equal to the third change amount, in step 118, a suspicion flag indicating that the flame may be erroneously determined by hot air is established. On the other hand, if the second variation is larger than the third variation, the process proceeds to step 120.

  In step 120, the fire determination unit 54 derives a linear function formula from the first change amount, the second change amount, and the third change amount calculated in step 106, and uses the derived linear function formula Calculated values for each of the change amount, the second change amount, and the third change amount are calculated.

  Then, in step 122, the fire determination unit 54 derives an approximate straight line from the second change amount and the third change amount calculated in step 106, and uses the derived approximate straight line to correspond to the first change amount. The calculated value is calculated as the noise amount.

  In step 124, the fire determination unit 54 determines the first change amount, the second change amount, and the third change amount according to the suspicion flag set in step 118 and / or the noise amount calculated in step 122. Acquire a threshold value of the determination ratio for.

  In the next step 126, the fire determination unit 54 determines the first change amount, the second change amount, and the third change amount calculated in step 106, and the first change amount and the second change amount calculated in step 120. Amount, the calculated value for each of the third change amount, the first change amount, the second change amount, the determination ratio for the third change amount is calculated, using the threshold obtained in step 124, the first Whether the determination ratio for the amount of change is equal to or greater than the corresponding threshold value, the determination ratio for the second amount of change is less than or equal to the corresponding threshold value, and the determination ratio for the third amount of change is less than or equal to the corresponding threshold value. To judge. If the determination ratio for the first change amount is less than the corresponding threshold value, the determination ratio for the second change amount is greater than the corresponding threshold value, or the determination ratio for the third change amount is greater than the corresponding threshold value. Determines that no flame is detected, and terminates the fire determination processing routine while continuing the current mode.

  On the other hand, the determination ratio for the first change amount is greater than or equal to the corresponding threshold value, the determination ratio for the second change amount is less than or equal to the corresponding threshold value, and the determination ratio for the third change amount is less than or equal to the corresponding threshold value. If there is, it is determined that the flame is detected, and the process proceeds to step 128.

  In step 128, the number-of-times determination unit 56 determines the number of times that the flame has been detected continuously based on the determination result of step 126 and the determination result of step 126 in the past. It is determined whether or not the above. If the number of times it is determined that the flame is continuously detected is the number of times N or more in succession, it is determined that a fire has occurred, and the process proceeds to step 134. On the other hand, when the number of times it is determined that the flame is continuously detected is less than the continuous number N, the process proceeds to step 130.

  In step 130, the number-of-times determination unit 56 determines in advance the number of times it is determined that the flame has been detected within a certain period of time, based on the determination result of step 126 and the past determination result of step 126. It is determined whether or not the number of accumulations is M or more. When the number of times that it is determined that the flame is detected within the fixed time is the cumulative number M or more, it is determined that a fire has occurred, and the process proceeds to step 134. On the other hand, when the number of times it is determined that the flame has been detected continuously is less than the cumulative number M, the process proceeds to step 132, the caution output mode is entered, and the fire determination processing routine is ended.

  In step 134, the number-of-times determination unit 56 determines the moving average value for each of the value of the electric signal from the first sensor 12, the value of the electric signal from the second sensor 14, and the value of the electric signal from the third sensor 16. Set to fix to the current moving average value. Then, in step 136, the alarm control unit 58 shifts to the fire mode, outputs a fire signal to the alarm control unit 46A and the alarm output unit 46B, and ends the fire determination processing routine. The fire signal output to the alarm output unit 46B activates the contact output forming the external output unit 32.

  As described above, according to the flame detector of the first embodiment, infrared light in the band near 4.5 μm of the carbon dioxide resonance radiation band and 4.0 μm in the wavelength band shorter than the carbon dioxide resonance radiation band. Each of the infrared light in the near band and the infrared light in the band near 5.0 μm of the wavelength band longer than the carbon dioxide resonance radiation band is detected and converted into an electric signal of a direct current component, Based on the first change amount, which is the change amount from the moving average value, the second change amount, and the third change amount, by determining whether or not a flame is detected, malfunction due to hot air is eliminated, and accuracy is improved. Can detect flames well.

  Further, since a thermopile that detects each infrared light and converts it into an electric signal of a DC component is used, frequency decomposition processing is unnecessary and early detection is possible.

In the first embodiment, the corrected sensor value of the electric signal from the first sensor 12, the corrected sensor value of the electric signal from the second sensor 14, and the corrected third sensor 16 are used. The threshold value E indicating the minimum detection sensitivity may be changed according to the sensor value of the electric signal of.
For example, the fire determination unit 54 detects the sensor value of the electric signal from the first sensor 12, the sensor value of the electric signal from the second sensor 14, and the electric signal from the third sensor 16 corrected by the correction unit 44. If at least one of the sensor values is outside the predetermined range, the threshold value E used in the fire determination is changed to be high.

  Here, the principle of changing the threshold value E representing the minimum detection sensitivity will be described with reference to FIG.

  When a sudden environmental change that does not accompany a flame occurs, as a measure to prevent malfunction (problem when the moving average approaches the live value) due to environmental change noise, within the range defined based on the bias value. If it is a live value, it is set to high sensitivity (standard sensitivity), and if it is out of the range, it is set to low sensitivity. The sensitivity is changed by changing the minimum detection sensitivity (threshold value E). It should be noted that the range that serves as the reference for switching the sensitivity may be determined in consideration of the influence of the correction using the correction coefficient.

  In this way, when a sudden environmental change that does not accompany a flame occurs, the threshold E corresponding to the minimum detection sensitivity is changed to a low sensitivity to eliminate malfunction due to environmental change noise, and more accurately. The flame can be detected.

<Second Embodiment>
<System configuration>
Next, the flame detector according to the second embodiment will be described. In addition, about the part which becomes the same structure as 1st Embodiment, the same code | symbol is attached | subjected and description is abbreviate | omitted.

  In the second embodiment, the value of the electric signal from the first sensor 12, the value of the electric signal from the second sensor 14, and the third sensor 16 are calculated without calculating the moving average or the amount of change from the moving average. The difference from the first embodiment is that the value of the electric signal from the device is compared to determine whether or not the flame is detected, and that the correction unit is not provided.

  As shown in FIG. 13, the second arithmetic processing unit 30 of the flame detector according to the second embodiment includes a fire determination unit 54, a frequency determination unit 56, an alarm control unit 58, and an alarm output unit 46B. ing.

  Here, the principle of detecting a flame will be described.

  From Stefan Boltzmann's law, it is known that the radiant energy for each wavelength of a black body is as shown in FIG. The peaks for each temperature radiation follow Wien's displacement law. For example, when comparing the energies of 4.0 μm and 5.0 μm, it can be seen that the low temperature radiation rises to the right and the high temperature radiation rises to the left.

  For example, as shown in FIG. 14, the slopes of the energy values of 4.0 μm and 5.0 μm are different between the gas at 1000 ° C. (high temperature) and the gas at 100 ° C. (low temperature) that have peaks in the carbon dioxide resonance radiation band.

  Therefore, it is possible to discriminate between high temperature and low temperature by comparing the slopes of the approximate straight lines of the signal values of the infrared sensor (with a bandpass filter) that captures the vicinity of 4.0 μm and the vicinity of 5.0 μm. That is, if it goes up to the right, it can be judged to be low temperature, and if it goes up to the left, it can be judged to be high temperature.

  Next, if the signal value of the infrared sensor that captures the vicinity of 4.5 μm (with a band-pass filter) is higher than the calculated value of the straight line connecting the signal values of 4.0 μm and 5.0 μm, it can be determined as a fire. However, some flames do not emit high-temperature heat (such as LNG that does not emit smoke), so rather than simply comparing the slopes of the signal values of 4.0μm and 5.0μm, the flame is judged according to the slope. The threshold for the calculated value of 4.5 μm is changed (see FIG. 15).

  That is, when the signal value in the band near 4.0 μm is larger than the signal value in the band near 5.0 μm, the signal value in the band near 4.5 μm than the calculated value of the straight line connecting the values of 4.0 μm and 5.0 μm. Is larger than the threshold A, it is determined to be a flame. When the signal value in the band near 4.0 μm is less than or equal to the signal value in the band near 5.0 μm, the signal value in the band near 4.5 μm is smaller than the calculated value of the straight line connecting the values of 4.0 μm and 5.0 μm. When the threshold value is equal to or higher than B, it is judged as a flame. However, the relationship between the threshold A and the threshold B is threshold A <threshold B (FIG. 16).

  The threshold value is also changed according to the strength of the entire signal (see FIG. 17).

  According to the principle described above, in the present embodiment, the fire determination unit 54 causes the value of the electric signal from the first sensor 12, the value of the electric signal from the second sensor 14, and the electric signal from the third sensor 16. Is greater than or equal to the threshold value E, and the calculated value of the electric signal from the first sensor 12 obtained from the approximate straight line of the value of the electric signal from the second sensor 14 and the value of the electric signal from the third sensor 16. For the result of comparison with the threshold value A or the threshold value B, it is determined that the flame is detected when the predetermined condition is satisfied. The threshold A and the threshold B are examples of determination thresholds.

In addition, the value of the electric signal detected by the first sensor 12 is obtained from the approximate straight line obtained from the value of the electric signal detected by the second sensor 14 and the value of the electric signal detected by the third sensor 16. A corresponding calculated value is calculated as a noise amount, and a table of threshold _values A ₁ , threshold _values A ₂ , ..., Threshold value A _n determined according to the calculated noise amount is used.

  Here, when the value of the electrical signal from the second sensor 14 is larger than the value of the electrical signal from the third sensor 16, the fire determination unit 54 uses the threshold value A smaller than the threshold value B to perform the determination. On the other hand, when the value of the electric signal from the second sensor 14 is less than or equal to the value of the electric signal from the third sensor 16, the fire determination unit 54 uses a threshold value B that is larger than the threshold value A for severe determination, Make a decision.

<Operation of flame detector>
Next, the operation of the flame detector according to the second embodiment will be described. The same processes as those in the first embodiment are designated by the same reference numerals, and detailed description thereof will be omitted.

  A fire determination processing routine according to the second embodiment will be described with reference to FIGS. 18 and 19.

  In step 100, the signal acquisition unit 40 determines the value of the electric signal from the first sensor 12, the value of the electric signal from the second sensor 14, the electric value from the third sensor 16 from the signal output from the AD conversion unit 26. Get the value of a signal.

  Then, in step 300, the fire determination unit 54 determines the value of the electric signal from the first sensor 12, the value of the electric signal from the second sensor 14, and the value of the electric signal from the third sensor 16 acquired in step 100. Are all less than a predetermined threshold value E. When it is determined that the value of the electric signal from the first sensor 12, the value of the electric signal from the second sensor 14, and the value of the electric signal from the third sensor 16 are all less than a predetermined threshold value E. In step 114, if the current mode is the fire mode or the caution output mode, the normal mode is entered, and the fire determination processing routine is ended. If the current mode is the normal mode, the normal mode is continued.

  It is determined that at least one of the value of the electric signal from the first sensor 12, the value of the electric signal from the second sensor 14, and the value of the electric signal from the third sensor 16 is equal to or greater than a predetermined threshold value E. In that case, in step 302, the fire determination unit 54 determines whether the value of the electric signal from the second sensor 14 is less than or equal to the value of the electric signal from the third sensor 16. If the value of the electrical signal from the second sensor 14 is less than or equal to the value of the electrical signal from the third sensor 16, in step 118, the suspicion flag indicating that the flame may be erroneously determined by hot air is turned on. To On the other hand, when the value of the electric signal from the second sensor 14 is larger than the value of the electric signal from the third sensor 16, the process proceeds to step 120.

  In step 304, the fire determination unit 54 derives an approximate straight line from the value of the electric signal from the second sensor 14 and the value of the electric signal from the third sensor 16 acquired in step 100, and obtains the derived approximate straight line. The calculated value for the value of the electric signal from the first sensor 12 is calculated using the calculated value. The calculated value is also used as the noise amount.

  In step 306, the fire determination unit 54 sets the threshold value A as the threshold value related to the value of the electric signal from the first sensor 12 according to the suspicion flag set in step 118 and the calculation amount calculated in step 304. Alternatively, the threshold B is acquired. Here, the threshold A and the threshold B are set for each calculation amount, and the threshold A and the threshold B are set to increase as the calculation amount increases. Further, the threshold value A and the threshold value B are set so that the threshold value A <the threshold value B for the same calculation amount.

  In the next step 308, the fire determination unit 54 calculates the value of the electric signal from the first sensor 12 acquired in step 100 and the calculated value for the value of the electric signal from the first sensor 12 calculated in step 304. On the basis of the above, the suspicious flag in step 118 is not established, and the threshold value A or the threshold value B acquired in step 306 is used to determine the calculated value for the value of the electric signal from the first sensor 12. It is determined whether the difference is equal to or greater than the threshold A or the threshold B. When the difference between the calculated value and the value of the electric signal from the first sensor 12 is less than the threshold A or the threshold B, it is determined that the flame is not detected, and the fire determination processing routine is continued while the current mode is continued. To finish.

  On the other hand, when the difference between the value of the electric signal from the first sensor 12 and the calculated value with respect to the value of the electric signal from the first sensor 12 is equal to or larger than the threshold value A, it is determined that flame is detected, and the step Move to 128. On the other hand, when the difference between the value of the electric signal from the first sensor 12 and the calculated value with respect to the value of the electric signal from the first sensor 12 is less than the threshold value A, the fire is continued while the current mode is continued. The determination processing routine ends.

  In step 128, the number-of-times determination unit 56 determines the number of times that the flame has been detected continuously based on the determination result of step 126 and the determination result of step 126 in the past. It is determined whether or not the above. When the number of times it is determined that the flame has been continuously detected is equal to or more than the continuous number N, it is determined that a fire has occurred, and the process proceeds to step 136. On the other hand, when the number of times it is determined that the flame is continuously detected is less than the continuous number N, the process proceeds to step 130.

  In step 130, the number-of-times determination unit 56 determines in advance the number of times it is determined that the flame has been detected within a certain period of time, based on the determination result of step 126 and the past determination result of step 126. It is determined whether or not the number of accumulations is M or more. When the number of times it is determined that the flame is detected within the fixed time is the cumulative number M or more, it is determined that a fire has occurred, and the process proceeds to step 136. On the other hand, when the number of times it is determined that the flame has been detected continuously is less than the cumulative number M, the process proceeds to step 132, the caution output mode is entered, and the fire determination processing routine is ended.

  Then, in step 136, the alarm control unit 58 shifts to the fire mode, outputs a fire signal to the external output unit 32, and ends the fire determination processing routine.

  As described above, according to the flame detector of the second embodiment, infrared light in the band near 4.5 μm of the carbon dioxide resonance radiation band and 4.0 μm in the wavelength band shorter than the carbon dioxide resonance radiation band. Infrared light in the near band and infrared light in the band near 5.0 μm, which is a longer wavelength band than the carbon dioxide resonance radiation band, are detected and converted into electric signals of DC component, and the value of each electric signal is detected. By comparing with each other to determine whether or not the flame is detected, the malfunction due to the hot air can be eliminated, and the flame can be accurately detected.

  The present invention is not limited to the above-described embodiments, and various modifications and applications are possible without departing from the scope of the present invention.

  For example, in the above-described first embodiment, a case where it is determined whether or not a flame is detected using a threshold for strict determination when the third variation is larger than the second variation is described as an example. However, the present invention is not limited to this. Only when the third change amount is smaller than the second change amount, it may be determined that the flame is detected if the condition regarding the determination ratio is satisfied. In this case, if the third change amount is larger than the second change amount, it is not determined that the flame is detected.

  Moreover, in the said 1st Embodiment-2nd Embodiment, when the number of times that it was determined by the fire determination part 54 that the flame was detected continuously is more than the predetermined continuous number of times, or the fire determination part 54. The case where the fire signal is output when the number of times that it is determined that the flame is detected within a predetermined time is equal to or greater than the predetermined cumulative number has been described as an example, but the present invention is not limited to this. It is determined that the number of times that the fire determination unit 54 has continuously detected flames is equal to or greater than a predetermined number of consecutive times, and that the fire determination unit 54 has detected flames within a certain period of time. The fire signal may be output when the number of times is equal to or greater than a predetermined cumulative number.

  Further, the case where infrared rays in the band near 4.0 μm and infrared rays in the band near 5.0 μm other than the band near 4.5 μm of the carbon dioxide resonance radiation band are detected has been described as an example, but the invention is not limited to this. . In addition to the band near 4.5 μm of the carbon dioxide resonance radiation band, if the infrared ray has a wavelength shorter than the carbon dioxide resonance radiation band and the infrared ray has a wavelength longer than the carbon dioxide resonance radiation band, the infrared rays of other bands May be detected, or infrared rays in two or more bands may be respectively detected as a wavelength band shorter than the carbon dioxide resonance radiation band, or as a wavelength band longer than the carbon dioxide resonance radiation band, Infrared rays in two or more bands may be detected respectively.

  Further, the average calculator 50 has been described by taking as an example the case where the moving average value of the values of the electric signals from the respective sensors is calculated as the monitoring environment signal value of the signals of the respective sensors, but the present invention is not limited to this. The weighted average value of the values of the electric signals from the respective sensors may be calculated as the monitoring environment signal value of the signals of the respective sensors.

  Further, in the second embodiment, the first sensor 12 obtained from the approximate straight line obtained from the value of the electric signal detected by the second sensor 14 and the value of the electric signal detected by the third sensor 16. Although the calculation value corresponding to the value of the electric signal detected by is calculated as the amount of noise has been described as an example, the present invention is not limited to this, and the value of the electric signal detected by the first sensor 12, It corresponds to the value of the electric signal detected by the first sensor 12, which is obtained from the approximate straight line obtained from the value of the electric signal detected by the second sensor 14 and the value of the electric signal detected by the third sensor 16. The calculated value may be calculated as the amount of noise.

  Further, in the first embodiment, the case where the threshold value is acquired from the threshold value table has been described as an example, but a function for obtaining the threshold value may be derived and the threshold value may be acquired from the function. .

  Further, in the second embodiment, as in the first embodiment, the correction unit calculates the value of the electric signal detected by the first sensor 12 and the electric signal detected by the second sensor 14. The sensor value of the electric signal detected by the first sensor 12 is corrected by using the correction coefficient set by the correction coefficient setting unit 42 with respect to the value and the value of the electric signal detected by the third sensor 16. The fire determination may be performed using the sensor value of the electric signal detected by the second sensor 14 and the sensor value of the electric signal detected by the third sensor 16.

  Further, in the second embodiment, similarly to the first embodiment, the average calculation unit 50 causes the moving average value of the sensor values of the electric signal from the first sensor 12 and the second sensor 14 to obtain the moving average value. The moving average value of the sensor value of the electric signal and the moving average value of the sensor value of the electric signal from the third sensor 16 may be calculated. Further, the difference between the live value of the electric signal from the first sensor 12 and the moving average value of the signal of the first sensor 12 calculated by the average calculating section 50 by the change amount calculating section 52 is set as the first change amount. The difference between the live value of the electric signal from the second sensor 14 and the moving average value of the signal of the second sensor 14 calculated by the average calculating unit 50 is calculated as the second change amount, and the third sensor is calculated. The difference between the live value of the electric signal from 16 and the moving average value of the signal of the third sensor 16 calculated by the average calculating unit 50 may be calculated as the third change amount. In this case, similarly to the first embodiment, the fire determination may be performed using the first change amount, the second change amount, and the third change amount. Further, as in the case of the above-described first embodiment, the moving average value may be reset when the change amount of the signal becomes smaller than the negative reference value.

  In addition, in the second embodiment, the fire determination unit 54 causes the correction unit 44 to correct the sensor value of the electrical signal from the first sensor 12, the sensor value of the electrical signal from the second sensor 14, and When at least one of the sensor values of the electric signal from the third sensor 16 is outside the predetermined range, the threshold value E used in the fire determination may be increased.

10 Flame detector 12 1st sensor 12A, 14A, 16A Filter 12B, 14B, 16B Detection element 14 2nd sensor 16 3rd sensor 18, 20, 22, 24 Amplification part 26 AD conversion part 28 1st arithmetic processing part 30 Second arithmetic processing unit 32 External output unit 40 Signal acquisition unit 42 Correction coefficient setting unit 44 Correction unit 46A Alarm display unit 46B Alarm output unit 50 Average calculation unit 52 Change amount calculation unit 54 Fire determination unit 56 Number of times determination unit 58 Alarm control Department

Claims (9)

A first bandpass filter that transmits infrared light including the peak wavelength of the carbon dioxide resonance radiation band;
A second band-pass filter that transmits infrared light in a band including a band of a wavelength shorter than the carbon dioxide resonance radiation band and has a band center at a position apart from the band center of the carbon dioxide resonance radiation band;
A third band-pass filter that transmits infrared light in a band including a band having a wavelength longer than that of the carbon dioxide resonance radiation band and has a band center at a position apart from the band center of the carbon dioxide resonance radiation band;
A first detection element that detects infrared light transmitted through the first band-pass filter and converts the infrared light into an electric signal;
A second detection element for detecting infrared light transmitted through the second band filter and converting the infrared light into an electric signal;
A third detection element for detecting infrared light transmitted through the third band filter and converting the infrared light into an electric signal;
A first average value of the electrical signal detected by the first detection element, a second average value of the electrical signal detected by the second detection element, and a third average of the electrical signal detected by the third detection element. An average calculation unit for calculating values,
A first change amount of the electric signal detected by the first detection element from the first average value, a second change amount of the electric signal detected by the second detection element from the second average value, And a change amount calculation unit that calculates a third change amount of the electric signal detected by the third detection element from the third average value,
A noise amount is calculated based on the second change amount and / or the third change amount calculated by the change amount calculation unit, and the determination threshold value is set so that the determination threshold value decreases as the calculated noise amount increases. Change, the ratio or difference between the first change amount and the noise amount, when the determination threshold is larger than the determination threshold, a determination unit that determines that a flame is detected ,
Flame detector including . The determination unit obtains an approximate straight line representing the relationship between the wavelength and the change amount from the second change amount and the third change amount calculated by the change amount calculation unit, and obtains from the approximate straight line, the carbonic acid The flame detector according to claim 1 , wherein the amount of change corresponding to the peak wavelength of the gas resonance radiation band is calculated as the amount of noise. The determination unit changes the determination threshold based on whether the second change amount is larger than the third change amount and determines whether or not a flame is detected. The flame detector described. When the second change amount is larger than the third change amount , the determination unit uses the first threshold value as the determination threshold value to determine whether or not flame is detected,
When the second variation amount is equal to or less than the third variation amount , a second threshold value larger than the first threshold value is used as the determination threshold value to determine whether or not flame is detected. Flame detector. The determination unit, from the first change amount calculated by the change amount calculation unit, the second change amount, and the third change amount , to obtain an approximate straight line representing the relationship between the wavelength and the change amount, the approximation The flame detector according to claim 1 , wherein a variation amount corresponding to a peak wavelength of the carbon dioxide resonance radiation band, which is obtained from a straight line, is calculated as a noise amount. Or the determination unit, the first change amount calculated by the change amount calculating unit, the second change amount, and the third change amount of at least one, but if it is above the detection threshold, and detects a flame The flame detector according to any one of claims 1 to 5, which determines whether or not the flame detector is used. Further including a number of times determination unit,
The determination unit repeatedly determines whether or not a flame is detected in a constant cycle,
The number of times determination unit, if the number of times it is determined that the flame is continuously detected is equal to or more than a predetermined continuous number of times, and / or the number of times that it is determined that the flame is detected within a certain period is predetermined. The flame detector according to any one of claims 1 to 6, which outputs a fire signal when the accumulated number of times is equal to or more than the accumulated number of times. The first detection element detects infrared light transmitted through the first band-pass filter and converts the infrared light into a DC component electric signal,
The second detection element detects the infrared light transmitted through the second band filter and converts the infrared light into a DC component electric signal,
The flame detector according to any one of claims 1 to 7, wherein the third detection element detects infrared light that has passed through the third band filter and converts the infrared light into a DC component electric signal.   The flame detector according to claim 8, wherein a thermopile or an InAsSb element is used for the first detection element, the second detection element, and the third detection element.