Fire sign detection system, fire sign detection method, and fire sign detection program

Abstract

To provide a fire sign detection system, a fire sign detection method and a fire sign detection program that are able to detect a sign of a fire in an early stage.SOLUTION: A fire sign detection system capable of communicating with a plurality of types of sensors installed in an area to be monitored and configured to measure an environment of the area and for detecting a sign of a fire in the area, includes: a measurement data acquisition unit configured to acquire measurement data of each of a plurality of indices of the environment of the area acquired by the plurality of types of sensors; a sign presence / absence determination unit configured to determine presence or absence of the sign of the fire in the area, based on a combination of measurement data of a predetermined index selected from the measurement data acquired by the measurement data acquisition unit; and a notification unit configured to notify a person monitoring the area of the fact that there is the sign of the fire, when the sign presence / absence determination unit determines that there is the sign of the fire in the area.SELECTED DRAWING: Figure 5

Description

[Technical field]

[0001] The present invention relates to a fire sign detection system, a fire sign detection method, and a fire sign detection program, and more particularly to a fire sign detection system, a fire sign detection method, and a fire sign detection program that can detect a fire at an early stage. [Background technology]

[0002] Buildings and other structures are equipped with firefighting equipment such as fire alarms and sprinklers, but at construction sites of buildings under construction, fire alarms and firefighting equipment may not be installed, and the only fire prevention measures available may be security guards patrolling the site and buckets of water for extinguishing the fire.

[0003] For example, at construction sites where buildings are under construction, fires have occurred due to cigarette fires, sparks from welding, flying molten metal, etc. When people are present, these fires tend to smolder and go unnoticed, but when people leave the building a few hours later, flames break out and the fire spreads. For these reasons, it is especially important to detect fires early at construction sites, and there is a demand for equipment that can detect early signs of fire, such as smoldering fires. In addition, outdoor and open-air facilities may not be equipped with fire detection equipment, and fires caused by careless handling of fires during the daytime when people are present may break out a few hours later when no one is there. For this reason, outdoor and open-air facilities are also required to have equipment that can detect early signs of fire.

[0004] Patent Document 1 proposes a fire detection system that detects the presence or absence of smoke in the air. Since the amount of smoke is small in a smoldering state that indicates a fire, the fire detection system disclosed in Patent Document 1 may have difficulty detecting the smoldering state, and there is a risk that the fire signs will not be discovered. There is still a demand for a fire sign detection system that can detect fire signs such as a smoldering state at an early stage. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Patent Publication No. 2023-075361 Summary of the Invention [Problem to be solved by the invention]

[0006] Therefore, an object of the present invention is to provide a fire sign detection system, a fire sign detection method, and a fire sign detection program that are capable of detecting signs of a fire at an early stage. [Means for solving the problem]

[0007] In other words, the fire warning detection system of the first aspect is a fire warning detection system that is installed in an area to be monitored, is capable of communicating with multiple types of sensors that measure the environment of the area, and detects warning signs of a fire in the area, and is characterized by comprising a measurement data acquisition unit that acquires measurement data for each of multiple indicators of the area's environment acquired by the multiple types of sensors, a warning sign determination unit that determines whether or not there are warning signs of a fire in the area based on a combination of measurement data for predetermined indicators selected from the measurement data acquired by the measurement data acquisition unit, and an alarm unit that alerts a monitor of the area that there are warning signs of a fire when the warning sign determination unit determines that there are warning signs of a fire in the area.

[0008] In a second aspect, in the fire warning detection system according to the first aspect, the warning presence / absence determination unit may use a warning learning model that has pre-learned the correspondence between combinations of measurement data of specified indicators and the presence or absence of warning signs of a fire in the area, and determine the presence or absence of warning signs of a fire in the area by inputting into the warning learning model combinations of measurement data of specified indicators selected from the measurement data acquired by the measurement data acquisition unit.

[0009] In a third aspect, in the fire warning detection system of the first aspect, a reception unit is further provided that receives the type of area to be monitored, and the warning presence / absence determination unit may refer to a table that defines the correspondence between the type of area to be monitored and combinations of measurement data of indicators, and determine the combination of measurement data of indicators based on the type of area to be monitored received by the reception unit.

[0010] A fourth aspect is a fire warning detection system according to the first aspect, further comprising a risk judgment unit that, when the warning presence / absence judgment unit judges that there are warning signs of a fire in an area, judges the risk level of the warning signs based on a combination of measurement data of specified indicators on which the warning presence / absence judgment unit based its judgment, and the notification unit may, when the warning presence / absence judgment unit judges that there are warning signs of a fire in an area, notify a monitor of the area that there are warning signs of a fire and the risk level of the warning signs.

[0011] In a fifth aspect, in the fire warning detection system according to the fourth aspect, the danger assessment unit may use a danger learning model that has previously learned the correspondence between combinations of measurement data of specified indicators and the danger of fire warning signs in an area, and judge the danger of fire warning signs in an area by inputting into the danger learning model the combination of measurement data of specified indicators that formed the basis for the warning sign presence / absence determination unit's determination that there are warning signs of a fire in the area.

[0012] In a sixth aspect, in the fire precursor detection system according to the first aspect, the measurement data acquisition unit may be configured so that one or more thresholds are set for the measurement data of each of the multiple indicators, and measurement data that exceeds the threshold for each of the indicators may be acquired from each of the multiple types of sensors.

[0013] According to a seventh aspect, in the fire sign detection system according to the first aspect, the multiple types of sensors may include an odor sensor that measures an odor in the area.

[0014] In an eighth aspect, in the fire warning detection system of the first aspect, the multiple types of sensors may include a two-dimensional thermopile radiation temperature sensor that divides a two-dimensional plane of the area into predetermined sections and measures a two-dimensional temperature distribution that numerically indicates the temperature of each section.

[0015] As for a ninth aspect, in the fire sign detection system according to the first aspect, the plurality of types of sensors may include a barometric pressure sensor that measures a barometric pressure in the area.

[0016] According to a tenth aspect, in the fire sign detection system according to the first aspect, the plurality of types of sensors may include a temperature and humidity sensor that measures temperature and humidity in a steady state of the area.

[0017] According to an eleventh aspect, in the fire warning detection system according to the first aspect, the plurality of types of sensors may include an air pollution sensor that measures air pollution in the area.

[0018] According to a twelfth aspect, in the fire sign detection system according to the first aspect, the plurality of types of sensors may include a smoke sensor that detects a concentration of smoke in an area.

[0019] According to a thirteenth aspect, in the fire sign detection system according to the first aspect, the plurality of types of sensors may include a carbon monoxide sensor that detects a concentration of carbon monoxide in the area.

[0020] According to a fourteenth aspect, in the fire sign detection system according to the first aspect, the plurality of types of sensors may include a carbon dioxide sensor that detects a concentration of carbon dioxide in the area.

[0021] According to a fifteenth aspect, in the fire sign detection system according to the first aspect, the plurality of types of sensors may include an illuminance sensor that measures illuminance of an area.

[0022] According to a sixteenth aspect, in the fire sign detection system according to the first aspect, the plurality of types of sensors may include a noise sensor that measures the loudness of noise in an area.

[0023] According to a seventeenth aspect, in the fire sign inspection system according to the first aspect, the plurality of types of sensors may include an oxygen sensor that measures an oxygen concentration in the area.

[0024] In an 18th aspect, in the fire warning inspection system of the first aspect, the multiple types of sensors may be mounted on an unmanned aerial vehicle or robot, and the unmanned aerial vehicle or robot may fly or travel in an area by remote control or autonomous control.

[0025] In a 19th aspect, the fire warning inspection system according to the first aspect may further include an abnormal value extraction unit that performs unsupervised clustering of the measurement data acquired by the measurement data acquisition unit to extract outliers in the measurement data for each of a plurality of indicators of the environment of the area included in the measurement data as abnormal values ​​of the measurement data for each of the plurality of indicators, and an abnormal value indicator determination unit that determines whether or not there are indicators of a fire in the area based on the abnormal values.

[0026] A twentieth aspect is a fire warning inspection system according to the sixth aspect, further comprising an abnormal value extraction unit that extracts outliers of the measurement data for each of a plurality of indexes of the environment of the area included in the measurement data as abnormal values ​​of the measurement data for each of the plurality of indexes by unsupervised clustering of the measurement data acquired by the measurement data acquisition unit, and a threshold resetting unit that resets a threshold for the measurement data for each of the plurality of indexes based on the abnormal values, and the measurement data acquisition unit may acquire measurement data from each of the plurality of types of sensors that exceeds the threshold for each of the indexes reset by the threshold resetting unit.

[0027] A fire warning detection method according to a 21st aspect is a fire warning detection method that is capable of communicating with multiple types of sensors that are installed in a monitored area and measure the environment of the area, and detects warning signs of a fire in the area, wherein the computer executes the following steps: a measurement data acquisition step that acquires measurement data for each of multiple indicators of the area's environment acquired by the multiple types of sensors; a warning sign determination step that determines whether or not there are warning signs of a fire in the area based on a combination of measurement data for predetermined indicators selected from the measurement data acquired in the measurement data acquisition step; and a notification step that notifies a monitor of the area that there are warning signs of a fire if it is determined that there are warning signs of a fire in the area in the warning sign determination step.

[0028] A fire warning detection program according to a 22nd aspect is a fire warning detection program that is installed in an area to be monitored, is capable of communicating with multiple types of sensors that measure the environment of the area, and detects warning signs of a fire in the area, and is characterized in that it has a computer perform: a measurement data acquisition function that acquires measurement data for each of multiple indicators of the area's environment acquired by the multiple types of sensors; a warning sign determination function that determines whether or not there are warning signs of a fire in the area based on a combination of measurement data for predetermined indicators selected from the measurement data acquired by the measurement data acquisition function; and an alarm function that alerts a monitor of the area that there are warning signs of a fire when the warning sign determination function determines that there are warning signs of a fire in the area. Effect of the Invention

[0029] The fire warning detection system etc. of the present invention is a fire warning detection system that is installed in a monitored area, is capable of communicating with multiple types of sensors that measure the environment of the area, and detects warning signs of a fire in the area, and is characterized by having a measurement data acquisition unit that acquires measurement data for each of multiple indicators of the area's environment acquired by the multiple types of sensors, a warning sign presence / absence determination unit that determines whether or not there are warning signs of a fire in the area based on a combination of measurement data for predetermined indicators selected from the measurement data acquired by the measurement data acquisition unit, and an alarm unit that alerts a monitor of the area that there are warning signs of a fire when the warning sign presence / absence determination unit determines that there are warning signs of a fire in the area, thereby making it possible to detect warning signs of a fire early. [Brief description of the drawings]

[0030] [Figure 1] FIG. 1 is a diagram for explaining an overview of a fire sign detection system according to this embodiment. [Diagram 2] FIG. 2 is an example of an infrastructure configuration diagram including a fire warning detection system according to this embodiment. [Diagram 3] FIG. 3 is a diagram for explaining an example of an installation mode of the environment measurement terminal according to this embodiment. [Figure 4] FIG. 4 is a block diagram for explaining an example of a hardware configuration of the fire sign detection system according to this embodiment. [Diagram 5] FIG. 5 is a block diagram for explaining an example of the hardware configuration of the environment measurement terminal according to this embodiment. [Figure 6a] FIG. 6a is a diagram for explaining an example of a circuit configuration of the environment measurement terminal according to this embodiment. [Figure 6b] FIG. 6b is a diagram for explaining an example of a circuit configuration of the environment measurement terminal according to this embodiment. [Figure 6c] FIG. 6c is a diagram for explaining an example of a circuit configuration of the environment measurement terminal according to this embodiment. [Figure 6d] FIG. 6d is a diagram for explaining an example of a circuit configuration of the environment measurement terminal according to this embodiment. [Figure 7]FIG. 7 is a block diagram for explaining an example of the functional configuration of the fire sign detection system according to this embodiment. [Figure 8] FIG. 8 is a diagram showing an example of a table that defines the correspondence between the type of area to be monitored according to this embodiment and combinations of measurement data of multiple indices of the environment of the area. [Figure 9] FIG. 9 is a diagram for explaining an example of a combination of measurement data of a plurality of indexes of the environment of an area to be monitored according to this embodiment. [Figure 10] FIG. 10 is a diagram for explaining an example of an experimental result of the fire sign detection system according to the present embodiment. [Figure 11] FIG. 11 is an example of a flowchart of a fire sign detection program according to this embodiment. [Figure 12] FIG. 12 is an example of a flowchart of a fire sign detection program according to another embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0031] (Overview of Fire Prediction and Detection System 10) An embodiment of a fire sign detection system 10 according to the present disclosure will be described with reference to FIGS. First, an overview of a fire sign detection system 10 according to this embodiment will be described with reference to FIG.

[0032] The fire warning detection system 10 is capable of communicating with multiple types of sensors that are installed in an area to be monitored and measure the environment of the area, and detects warning signs of a fire by measuring the environment of the area using the multiple types of sensors. The multiple types of sensors are mounted on an environment measuring terminal 20 described below. The multiple types of sensors are controlled by a CPU 20e of the environment measuring terminal 20, and measurement data of the multiple types of sensors is transmitted to the fire sign detection system 10.

[0033] By installing the environmental measurement terminal 20 in an area to be monitored, multiple types of sensors are installed in the area. Note that the multiple types of sensors are not limited to being built into the environmental measurement terminal 20, but may be installed separately in the area to be monitored away from the environmental measurement terminal 20, and the measurement data may be acquired by the environmental measurement terminal 20 and transmitted to the fire warning detection system 10. The area to be monitored is an area where it is desired to detect signs of fire, and construction sites of buildings, logistics warehouses, and various factories are assumed. The area to be monitored is not limited to construction sites, but may be any place where it is desired to monitor fires and detect signs of fires, and a wide range of areas may be the area to be monitored. For example, it may be a house, a nursing home, a childcare facility, a power plant, a forest, a dam, a fishing port, a railway line, an airport runway, etc., and may be either during construction work or after completion of construction work, and large-scale outdoor facilities such as fields, parks, grounds, and parking lots that do not involve buildings may also be the area to be monitored. Furthermore, the inside of a moving object such as a bus, a train, a ship, an airplane, a submarine, or a spaceship may also be the area to be monitored. If the area to be monitored exceeds the measurable range of multiple types of sensors implemented in one environmental measuring terminal 20, multiple environmental measuring terminals 20 are installed in the area to be monitored to cover the entire area to be monitored.

[0034] The multiple types of sensors in this embodiment refer to, as shown in FIG. 5, an odor sensor 21, a two-dimensional thermopile radiation temperature sensor 22, an air pressure sensor 23, a temperature and humidity sensor 24, an air pollution sensor 25, a smoke sensor 26, a carbon monoxide sensor 27, a carbon dioxide sensor 28, an illuminance sensor 29, a noise sensor 30, a camera with night vision function 31, and an oxygen sensor 32, and various environmental indicators of the area to be monitored are measured using these sensors. Furthermore, due to improvements in the fire warning detection system 10 or the environmental measuring terminal 20, some of these types of sensors may no longer be used and may no longer be installed in the environmental measuring terminal 20, or may be replaced with other sensors.

[0035] The odor sensor 21 measures odors in the area to be monitored. Among many odors, the odor sensor 21 detects a "fishy odor" that occurs when a fire is smoldering (smoldering state) as a sign of a fire, measures its strength, and outputs it as a numerical value. The odor refers to a state in which odor substances are in the form of fine particles or gas. The principles of the odor sensor 21 include a semiconductor type, a quartz crystal oscillator type, an FET biosensor type, and an MSS type. A pulse voltage is applied to the odor sensor 21, and a sensor resistance value corresponding to the concentration of odor molecules is detected. An analog voltage generated across the sensor resistance of the odor sensor 21 is input to a CPU 20e of the environment measuring terminal 20, which will be described later.

[0036] The semiconductor method derives the concentration of odor molecules by measuring the resistance that changes when the odor molecules are adsorbed by oxygen on the semiconductor surface. Metal oxide semiconductors are the most common type of semiconductor, but there are also methods that use organic semiconductors. In the quartz crystal oscillator method, when the sensitive film attached to the quartz crystal oscillator captures odor molecules, the resonant frequency of the oscillator drops, and the concentration of the odor molecules is derived by measuring the amount of frequency drop. Natural or synthetic lipids are used for the sensitive film. When multiple sensors with sensitive films are used simultaneously, multiple odor molecules can be identified.

[0037] In the FET biosensor method, a probe molecule fixed on the insulating film of a transistor detects the charge of adsorbed gas molecules, and by analyzing the electrical signal caused by that charge, it is possible to derive the composition and concentration of the gas molecules. FET is an abbreviation for Field Effect Transistor. In the MSS method, stress is generated when gas molecules are adsorbed on the sensitive film, and the components and concentration of the gas molecules are derived by analyzing the electrical resistance that changes due to the stress. MSS is an abbreviation for Membrane-type Surface stress Sensor.

[0038] The two-dimensional thermopile radiation temperature sensor 22 measures a two-dimensional temperature distribution by dividing a two-dimensional plane of a monitoring target area into predetermined sections and indicating the temperature of each section as a numerical value. The two-dimensional thermopile radiation temperature sensor 22 is capable of measuring two-dimensional temperature distribution at a remote location. The two-dimensional thermopile radiation temperature sensor 22 measures two-dimensional heat distribution by collecting infrared radiation energy radiated from a remote object using a lens and irradiating it on a total of 64 thermopiles arranged in an 8×8 matrix. Therefore, the two-dimensional thermopile radiation temperature sensor 22 divides the monitoring target area into 64 sections arranged in an 8×8 matrix and measures two-dimensional temperature distribution by associating each of the temperatures of the sections with the detected temperatures of the 64 thermopiles.

[0039] The two-dimensional thermopile radiation temperature sensor 22 digitally outputs the measured temperature data in a format that responds to a data request from the CPU 20e of the environment measurement terminal 20 described below via I2C communication, and transmits the temperatures detected by the 64 thermopiles as 64 pieces of temperature data at predetermined intervals. I2C (Inter-Integrated Circuit) communication is one of the communication interfaces for synchronous serial communication that communicates data in synchronization with a clock signal. Note that the communication method of the two-dimensional thermopile radiation temperature sensor 22 is not limited to I2C communication, and can be changed to another communication method due to specification changes or improvements of the two-dimensional thermopile radiation temperature sensor 22.

[0040] This intermittent time can be switched by a rotary switch incorporated in a sensor circuit including the two-dimensional thermopile radiation temperature sensor 22, and the intermittent time can be switched to 4 second intervals, 10 second intervals, 30 second intervals, or 60 second intervals. Note that the intermittent time is not limited to these, and other intermittent times may be used. In addition, the intermittent time can be synchronized with the timing at which a low level of the TX signal in communication by the TX interface is detected, and the two-dimensional thermopile radiation temperature sensor 22 can output and transmit 64 pieces of temperature data at the timing at which a low level of the TX signal is detected.

[0041] The two-dimensional thermopile radiation temperature sensor 22 starts measurement when a measurement start push button switch incorporated in a sensor circuit including the two-dimensional thermopile radiation temperature sensor 22 is pressed.

[0042] The air pressure sensor 23 measures the air pressure in the area to be monitored. Since the output values ​​of the various sensors vary depending on the value of the atmospheric pressure, the output values ​​of the various sensors are corrected based on the measured value of the atmospheric pressure in the area to be monitored measured by the atmospheric pressure sensor 23 . The air pressure sensor 23 is an 8-pin DIP module, and the measured air pressure is internally corrected and converted to obtain the air pressure (hPa) by dividing the 24-bit reading by 4096. The measurement data of the air pressure sensor 23 is output digitally in a format that responds to a data request from the CPU 20e of the environment measurement terminal 20 (described later) via I2C communication. The communication method of the air pressure sensor 23 is not limited to I2C communication, and may be changed to another communication method due to changes or improvements in the specifications of the air pressure sensor 23.

[0043] The temperature and humidity sensor 24 measures the temperature and humidity of the area to be monitored. The temperature and humidity sensor 24 measures the temperature and humidity of the area in a steady state, making it possible to monitor changes in temperature and humidity due to smoldering of a fire. The temperature and humidity sensor 24 is a composite sensor module that simultaneously measures temperature and humidity, and outputs digital data in a format that responds to a data request from the CPU 20e of the environment measurement terminal 20 (described later) via I2C communication. The communication method of the temperature and humidity sensor 24 is not limited to I2C communication, and may be changed to another communication method due to changes or improvements in the specifications of the temperature and humidity sensor 24.

[0044] The air pollution sensor 25 measures the air pollution in the area to be monitored. The air pollution sensor 25 is capable of detecting carbon monoxide gas, hydrogen gas, ammonia gas, methane gas, ethanol gas, propane gas, and isobutane gas, which are believed to cause air pollution, but is unable to identify the type of gas detected and has no gas selectivity. Although the air pollution sensor 25 of the present embodiment does not have the ability to individually identify and detect multiple types of gas, a gas sensor that individually identifies and detects multiple types of gas to be detected may be used as the air pollution sensor. Alternatively, a different type of gas sensor may be used to detect each of the multiple types of gas to be detected.

[0045] The air pollution sensor 25 is used with a 5V DC voltage and outputs a measured value as an analog voltage. When the air pollution sensor 25 detects gas, it amplifies the output voltage according to the proportion of the gas. The heater of the air pollution sensor 25 consumes 25 to 35 mA, and the operation can be stopped using the EN pin to reduce the current consumption. The output terminal of the air pollution sensor 25 is connected to a 10x offset differential amplifier circuit via a buffer circuit. The output terminal of the 10x differential amplifier circuit is connected to the CPU 20e of the environmental measurement terminal 20 described later, and the analog voltage of the air pollution sensor 25 is input to the CPU 20e of the environmental measurement terminal 20 described later. The buffer circuit is used for the purpose of correcting the output voltage and signal strength of the air pollution sensor 25.

[0046] The smoke sensor 26 detects the concentration of smoke in the area being monitored. The smoke sensor 26 is an optical particulate sensor that detects fine particulate matter (PM2.5). Fine particulate matter (PM2.5) refers to extremely small particles suspended in the air with a particle size of 2.5 micrometers or less, and its components include carbon components, nitrates, sulfates, ammonium salts, as well as inorganic elements such as silicon, sodium, and aluminum.

[0047] The smoke sensor 26 detects carbon monoxide, carbon dioxide gas, hydrogen cyanide (hydrogen prussic acid gas), chlorine gas, and the like, which are the main components of smoke from a fire. The smoke sensor 26 irradiates the air flowing in through a through hole in the center with light from a light-emitting element (LED), detects the light scattered by minute particles (smoke components) with a light-receiving element (photodiode), and amplifies and outputs the output signal of the light-receiving element. The light from the light-emitting element is narrowed by a lens and a slit. The light-receiving element is also narrowed by a lens and a slit, just like the light-emitting part, to cut out unnecessary light as much as possible and efficiently receive the detection light. The point where these optical axes intersect is the detection area.

[0048] Even when there are no microparticles, the light receiving element receives unwanted light and generates an output voltage. An output current proportional to the amount of light received is obtained from the light receiving element, which is amplified by an amplifier circuit and output as an analog voltage (pulse output) proportional to the dust concentration. The output terminal of the smoke sensor 26 is connected to the input terminal of a peak hold circuit, and the output terminal of the peak hold circuit is connected to the input terminal of an LPF (Low Pass Filter) circuit. The output terminal of the LPF circuit is connected to the CPU 20e of the environmental measurement terminal 20 described below. A peak hold circuit is a circuit that holds a maximum value for a certain period of time or until it is reset.

[0049] The carbon monoxide sensor 27 detects the concentration of carbon monoxide in the area to be monitored. The carbon monoxide sensor 27 is selective and capable of measuring the absolute value of carbon monoxide. The carbon monoxide sensor 27 is an electrochemical gas sensor. Electrochemical gas sensors use the principle of converting a chemical reaction into current and voltage to generate an electrical signal proportional to the amount of the substance to be measured. The carbon monoxide sensor 27 has a linear relationship between the carbon monoxide concentration and the sensor output current, and can be used to quantitatively measure the carbon monoxide concentration by calibrating the sensor output current with gas of known carbon monoxide concentration.

[0050] The voltages of both terminals sandwiching the load resistor connected to the carbon monoxide sensor 27 are connected to the input terminals of a differential amplifier circuit. The output terminal of the differential amplifier circuit is connected to the input terminal of an LPF circuit, and the output terminal of the LPF circuit is connected to the CPU 20e of the environment measuring terminal 20 described later.

[0051] Carbon dioxide sensor 28 detects the concentration of carbon dioxide in the area being monitored. The carbon dioxide sensor 28 is an NDIR (non-dispersive infrared) type carbon dioxide sensor, integrated as a module, which operates on a 5V DC power supply and outputs an analog voltage PWM. The NDIR (non-dispersive infrared) type gas sensor detects gas by utilizing the phenomenon in which infrared rays of a specific wavelength are absorbed as a result of the emitted infrared rays inducing molecular vibrations in the target gas (carbon dioxide). The infrared transmittance (the ratio of the transmitted light intensity to the emitted light intensity from the radiation source) is determined by the concentration of carbon dioxide, which is the target gas. Carbon dioxide molecules have the property of absorbing wavelengths of 4.26 μm in the infrared range, and the higher the carbon dioxide concentration, the more infrared rays of 4.26 μm are absorbed, so the carbon dioxide concentration can be measured by measuring the intensity of the infrared rays. The output terminal of the PWM output of the carbon dioxide sensor 28 is connected to a CPU 20e of the environment measuring terminal 20, which will be described later.

[0052] The illuminance sensor 29 measures the illuminance of the area to be monitored. The illuminance sensor 29 enables early and highly sensitive measurement of fires at night. The illuminance sensor 29 is a phototransistor that amplifies and outputs a photocurrent generated by irradiated light. The illuminance sensor 29 is connected to a temperature compensation circuit, and output characteristics that change with the operating temperature (temperature drift) are corrected. The output terminal of the illuminance sensor 29 is connected to the CPU 20e of the environment measurement terminal 20, which will be described later. Temperature drift refers to the amount of fluctuation in the output characteristics due to changes in the ambient temperature.

[0053] The noise sensor 30 measures the noise level in the area to be monitored. The noise sensor 30 is an ECM (electret condenser microphone). The output voltage of the noise sensor 30 is amplified by an amplifier circuit and then converted into an effective value, which is one of the indices that represent the magnitude of an AC electrical signal, by an rms (root mean square) conversion circuit. The signal then passes through a 10 Hz LPF circuit and a level conversion circuit, and is input to the CPU 20e of the environment measurement terminal 20, which will be described later. The effective value is a value that represents the effective magnitude of an AC electrical signal, and is used to compare the magnitude of the output voltage of the noise sensor 30.

[0054] The night vision camera 31 captures an image of the area to be monitored. The night vision camera 31 is equipped with a night vision correction function or an infrared function so that it can take pictures even when the area is dark. The night vision correction function is a function that automatically amplifies the sensitivity even in dim places with few light sources. The infrared function is a function that visualizes infrared rays emitted from an object. An output terminal of the night-vision camera 31 is connected to a CPU 20e of the environment measuring terminal 20, which will be described later, and the photographed data of the night-vision camera 31 is input to the CPU 20e. The environmental measurement terminal 20 may be equipped with a thermo camera instead of the night vision camera 31, or may be equipped with a thermo camera in addition to the night vision camera 31. A thermo camera is a type of infrared camera that captures infrared rays emitted by a subject to measure the temperature of the subject, and is also called a thermography camera. All objects above absolute zero have the characteristic of emitting infrared rays according to their temperature, and an infrared camera is a general term for cameras that capture and photograph the infrared rays emitted by a subject. Therefore, the environmental measurement terminal 20 can measure the temperature of the area to be monitored and can detect abnormalities such as high temperature conditions in the area. The environmental measurement terminal 20 can also detect specific gases, such as hydrocarbon gases such as methane, carbon monoxide, carbon dioxide, and ammonia, and can detect the generation of these specific gases in the area to be monitored.

[0055] Oxygen sensor 32 measures the concentration of oxygen in the monitored area. The oxygen sensor 32 is used for the purposes of detecting oxygen deficiency in the area to be monitored and managing the oxygen concentration. Oxygen deficiency refers to a decrease in the proportion of oxygen in the air, and can have serious effects on human health and survival. Since oxygen aids combustion, increased oxygen concentrations in the monitored area increase the risk of fire. Specifically, as the oxygen concentration increases, the ignition temperature of a substance decreases, making it easier to ignite, so the oxygen concentration measured by the oxygen sensor 32 is taken into account when determining whether or not there are signs of a fire in the area monitored by the sign-of-fire determination unit 14. In addition, the warning sign learning model includes the measurement data of the oxygen concentration in the monitored area measured by the oxygen sensor 32 in a combination of measurement data of predetermined indicators, and pre-learns the correspondence between these combinations of measurement data and the presence or absence of warning signs of fire in the monitored area. In addition, as the oxygen concentration increases, the flame temperature rises and combustion speeds up, expanding the combustion range and increasing the possibility of explosion, thereby increasing the danger. Therefore, the oxygen concentration measured by oxygen sensor 32 is taken into account when determining the danger of fire signs in the area monitored by danger determination unit 15. In addition, the risk learning model includes the measurement data of the oxygen concentration in the monitored area measured by the oxygen sensor 32 in a combination of measurement data of predetermined indices, and pre-learns the correspondence between these combinations of measurement data and the risk of fire precursors in the monitored area. An output terminal of the oxygen sensor 32 is connected to a CPU 20e of the environment measuring terminal 20, which will be described later, and measurement data of the oxygen sensor 32 is input to the CPU 20e.

[0056] It is assumed that the measurement data of the above-mentioned multiple types of sensors are affected by various noises. For this reason, the multiple types of sensors may be driven by a power supply that is boosted and stabilized by a DC-DC converter. In addition, the measurement data of the multiple types of sensors may be passed through an interference suppression filter and an outlier removal filter immediately before being input to the CPU 20e of the environmental measurement terminal 20 described below. In addition, noise contained in the measurement data of the multiple types of sensors may be separated from the measurement data, and a deep learning algorithm may be trained to remove the noise contained in the measurement data of the multiple types of sensors using the trained deep learning algorithm.

[0057] (Infrastructure configuration including Fire Prediction Detection System 10) Next, an infrastructure configuration including the fire sign detection system 10 will be described with reference to Fig. 2. Fig. 2 is an example of an infrastructure configuration diagram including the fire sign detection system 10 according to this embodiment.

[0058] As shown in FIG. 2, the fire sign detection system 10 and the environmental measurement terminals 20 (environmental measurement terminal A 20a, environmental measurement terminal B 20b) are connected to the Internet 11. The fire sign detection system 10 and the environmental measurement terminals 20 (environmental measurement terminal A 20a, environmental measurement terminal B 20b) are capable of two-way communication via the Internet 11. The fire sign detection system 10 and the environmental measurement terminals 20 (environmental measurement terminal A 20a, environmental measurement terminal B 20b) may be directly connected by wired communication or wireless communication without going through the Internet 11. Wired communication refers to, for example, communication by a wired LAN (Local Area Network). Wireless communication refers to, for example, short-range wireless communication such as wireless LAN or Bluetooth (registered trademark).

[0059] For convenience, two environmental measurement terminals 20 (environmental measurement terminal A 20a and environmental measurement terminal B 20b) are shown in Figure 2, but this is not limited to this, and one environmental measurement terminal 20, or three or more environmental measurement terminals 20 may be connected to and used with the fire warning detection system 10. Also, for convenience, FIG. 2 shows one fire warning detection system 10, but this is not limited to this, and two or more fire warning detection systems 10 may be connected to one or more environmental measurement terminals 20 and used.

[0060] (Regarding the manner of installation of the environment measuring terminal 20) Next, an installation mode of the environment measuring terminal 20 will be described with reference to Fig. 3. Fig. 3 is a diagram for explaining an example of an installation mode of the environment measuring terminal 20 according to this embodiment. The environmental measurement terminal 20 may be installed in a monitored area in a state where it is suspended from a ceiling 33, or may be installed by being attached to a stand 34 erected on the floor, or may be fixed to a worker's helmet 37. The environmental measurement terminal 20 may also be installed by being attached to a ceiling, wall, or floor, or may be attached to a device capable of moving in the air or on the ground, such as a drone, a robot, or a nanobot, so that it can be used while moving.

[0061] The multiple types of sensors may be mounted on an unmanned aerial vehicle or robot, and the unmanned aerial vehicle or robot may fly or run in the area to be monitored by remote control or autonomous control. 3, an environmental measurement terminal 20 is mounted on an unmanned aerial vehicle (UAV) 35 such as a drone, and multiple types of sensors mounted on the environmental measurement terminal 20 measure the environment of the area to be monitored. While the unmanned aerial vehicle 35 circles or hovers within the area to be monitored, multiple types of sensors mounted on the environmental measurement terminal 20 measure the environment of the area to be monitored. In addition to drones, the unmanned aerial vehicle 35 is also called a quadcopter or a multicopter. The unmanned aerial vehicle 35 may be remotely controlled by radio, or may fly using an automatic flight function based on autonomous control. Remote control by radio means that a person piloting the unmanned aerial vehicle 35 uses a controller (not shown) to pilot the unmanned aerial vehicle 35 from a remote location via wireless communication. The automatic flight function based on autonomous control means that the unmanned aerial vehicle 35 recognizes the surrounding environment by itself and flies according to a preset route or instructions, and flies according to a flight plan that has been created and implemented in advance. Measurement data from multiple types of sensors implemented in an environmental measurement terminal 20 installed on an unmanned aerial vehicle 35 is transmitted to the fire warning detection system 10 via wireless communication such as Wi-Fi (registered trademark) or Bluetooth.

[0062] 3, as an example of a robot, an environment measurement terminal 20 is installed on a four-legged robot 36, and multiple types of sensors mounted on the environment measurement terminal 20 measure the environment of the area to be monitored. While the four-legged robot 36 is stopped or running at a low speed within the area to be monitored, the multiple types of sensors mounted on the environment measurement terminal 20 measure the environment of the area to be monitored. The four-legged robot 36, like the unmanned aerial vehicle 35, may be remotely controlled by radio, or may travel using an automatic travel function based on autonomous control. Measurement data from multiple types of sensors implemented in the environmental measurement terminal 20 installed on the four-legged robot 36 is transmitted to the fire warning detection system 10 via wireless communication such as Wi-Fi (registered trademark) or Bluetooth, similar to the unmanned aerial vehicle 35. The robot is not limited to a four-legged robot 36, but may be a two-legged robot or a robot with three or more legs, a robot equipped with tires or wheels instead of legs, or a robot that does not have legs, tires, or wheels and moves using other means of movement (e.g., rails, ropes, etc.).

[0063] (Hardware configuration of the Fire Prediction Detection System 10) Next, a hardware configuration of the fire sign detection system 10 will be described with reference to Fig. 4. Fig. 4 is a block diagram for explaining an example of a hardware configuration of the fire sign detection system according to this embodiment.

[0064] The fire warning detection system 10 is a so-called computer, and includes a communication interface 10a, a read only memory (ROM) 10b, a random access memory (RAM) 10c, a storage unit 10d, a central processing unit (CPU) 10e, and an input / output interface 10f.

[0065] The communication interface 10a has a function of transmitting and receiving data handled by the fire warning detection system 10 to and from other devices via the Internet 11. The other devices are the environment measurement terminal 20 and other fire warning detection systems 10. When the fire warning detection system 10 connects to other devices via Bluetooth, the communication interface 10a becomes a Bluetooth interface. When the fire warning detection system 10 connects to other devices via wireless LAN, the communication interface 10a becomes a wireless LAN interface.

[0066] The storage unit 10d can be used as a storage device for the fire sign detection system 10, and is configured, for example, by a hard disk drive, a solid state drive, or a flash memory, and the storage unit 10d can also be configured using cloud storage. The storage unit 10d of the fire sign detection system 10 may be configured as a database. By configuring the storage unit 10d as a database, it is possible to efficiently manage a huge amount of data, and it is possible to quickly refer to required information with excellent searchability and accessibility.

[0067] In addition, the memory unit 10d of the fire warning detection system 10 stores the fire warning detection program described below, an OS (Operating System) required for the operation of the fire warning detection system 10, various other applications, and various data used by the applications.

[0068] The fire sign detection system 10 stores a fire sign detection program, which will be described later, in the ROM 10b or the storage unit 10d, and loads the fire sign detection program into a main memory constituted by the RAM 10c etc. The CPU 10e accesses the main memory into which the fire sign detection program has been loaded, and executes the fire sign detection program. The fire sign detection program may be stored in the ROM 20b or the storage unit 20d of each of the environment measuring terminals 20 and executed by the CPU 20e.

[0069] The input / output interface 10f transmits and receives data to and from devices external to the fire sign detection system 10. The external devices are an input device 10g and an output device 10h that input and output data to and from the fire sign detection system 10. The input device 10g is, for example, a keyboard and a mouse, and the output device 10h is, for example, a monitor, a printer, and a speaker.

[0070] (Hardware configuration of the environment measuring terminal 20) Next, the hardware configuration of the environment measuring terminal 20 will be described with reference to Fig. 5. Fig. 5 is a block diagram for explaining an example of the hardware configuration of the environment measuring terminal 20.

[0071] The environment measuring terminal 20 includes a communication interface 20a, a ROM 20b, a RAM 20c, a storage unit 20d, a CPU 20e, an input / output interface 20f, and a plurality of types of sensors. The plurality of types of sensors are directly connected to the CPU 20e. The plurality of types of sensors are an odor sensor 21, a two-dimensional thermopile radiation temperature sensor 22, an air pressure sensor 23, a temperature and humidity sensor 24, an air pollution sensor 25, a smoke sensor 26, a carbon monoxide sensor 27, a carbon dioxide sensor 28, an illuminance sensor 29, a noise sensor 30, and a camera with night vision function 31. Among the plurality of types of sensors, the measurement data of the two-dimensional thermopile radiation temperature sensor 22, the air pressure sensor 23, the temperature and humidity sensor 24, the camera with night vision function 31, and the oxygen sensor 32 are digitally output in a format that responds to a data request of the CPU 20e of the environment measuring terminal 20 described later by I2C communication. The measurement data of the carbon dioxide sensor 28 is input to the CPU 20e of the environment measuring terminal 20 as a PWM signal. In addition, the measurement data from the air pollution sensor 25, carbon monoxide sensor 27, illuminance sensor 29, and noise sensor 30 are passed through a differential amplification circuit, etc., and then the analog signals are converted into digital signals by a 12-bit ADC (Analog to digital converter) and input to the CPU 20e of the environmental measurement terminal 20. In addition, the measurement data of the smoke sensor 26 is peak-held for a certain period of time, and the analog signal is converted into a digital signal by a 12-bit ADC, and input to the CPU 20e of the environment measurement terminal 20. In addition, the communication methods of the odor sensor 21, the two-dimensional thermopile radiation temperature sensor 22, the air pressure sensor 23, the temperature and humidity sensor 24, the air pollution sensor 25, the smoke sensor 26, the carbon monoxide sensor 27, the carbon dioxide sensor 28, the illuminance sensor 29, the noise sensor 30, the camera with night vision function 31, and the oxygen sensor 32 may be changed to other communication methods due to changes or improvements in the specifications of these sensors.

[0072] The CPU 20e controls the odor sensor 21, the two-dimensional thermopile radiation temperature sensor 22, the air pressure sensor 23, the temperature and humidity sensor 24, the air pollution sensor 25, the smoke sensor 26, the carbon monoxide sensor 27, the carbon dioxide sensor 28, the illuminance sensor 29, the noise sensor 30, the camera with night vision function 31, and the oxygen sensor 32 to acquire measurement data from these multiple types of sensors and transmit it to the fire warning detection system 10.

[0073] The communication interface 20a has a function of transmitting and receiving data handled by the environment measuring terminal 20 to and from other devices via the Internet 11. The other devices are mainly the fire warning detection system 10. When the environment measuring terminal 20 connects to other devices via Bluetooth, the communication interface 10a becomes a Bluetooth interface. Also, when the environment measuring terminal 20 connects to other devices via wireless LAN, the communication interface 20a becomes a wireless LAN interface.

[0074] The storage unit 10d can be used as a storage device for the environment measuring terminal 20, and is composed of, for example, a hard disk drive, a solid state drive, a flash memory, etc. The storage unit 20d of the environment measuring terminal 20 stores firmware and various data required for the environment measuring terminal 20 to operate.

[0075] The input / output interface 20f transmits and receives data to and from devices external to the environment measuring terminal 20. The external devices refer to input and output devices (not shown) that input and output data to and from the environment measuring terminal 20.

[0076] The CPU 20e of the environmental measurement terminal 20 may be a PIC microcomputer (Peripheral Interface Controller micro computer). Although a PIC microcomputer is not suitable for general-purpose use such as a personal computer, it can be suitably used for control of specific applications such as control of peripheral devices such as sensors. Also, the CPU 20e may be a microcomputer such as M5stack, Raspberry Pi, Arduino, ESP, or various semiconductors equipped with device control functions.

[0077] (Circuit configuration of the environment measuring terminal 20) Next, a description will be given of a circuit configuration of the environment measuring terminal 20 with reference to Fig. 6a to Fig. 6d. Fig. 6a to Fig. 6d are diagrams for explaining an example of a circuit configuration of the environment measuring terminal 20.

[0078] FIG. 6a shows the configuration of analog circuits, mainly related to the air pollution sensor 25, the smoke sensor 26, and the carbon dioxide sensor 28, among the circuit configurations of the environment measuring terminal 20. FIG. 6b shows the configuration of an analog circuit mainly related to the carbon monoxide sensor 27 out of the circuit configuration of the environment measuring terminal 20. FIG. 6c shows, of the circuit configuration of the environment measuring terminal 20, the configuration of a digital circuit relating to the connection of terminals of the PIC microcomputer used mainly as the CPU 20e. FIG. 6d shows the configuration of digital circuits mainly related to the two-dimensional thermopile radiation temperature sensor 22, the air pressure sensor 23, the temperature and humidity sensor 24, and the illuminance sensor 29 in the circuit configuration of the environment measuring terminal 20.

[0079] (Functional configuration of the fire warning system 10) The functional configuration of the fire sign detection system 10 will be described with reference to Fig. 7. Fig. 7 is a block diagram for explaining an example of the functional configuration of the fire sign detection system 10.

[0080] By executing a fire precursor detection program described below, the fire precursor detection system 10 is provided with functional units such as a reception unit 12, a measurement data acquisition unit 13, a precursor presence / absence determination unit 14, a risk level determination unit 15, an alarm unit 16, an abnormal value extraction unit 17, an abnormal value precursor determination unit 18, and a threshold resetting unit 19 in the CPU 10e.

[0081] The receiving unit 12 receives the type of area to be monitored. The receiving unit 12 receives the type of area to be monitored, which is input by a user of the fire warning detection system 10 or an owner or supervisor of the area to be monitored. The types of areas to be monitored include, for example, buildings, logistics warehouses, petrochemical plants, machine factories, metal factories, food factories, pharmaceutical factories, paper factories, semiconductor factories, and other construction sites under construction, and the types are divided according to the materials placed therein. In addition, the types of areas to be monitored include houses, nursing homes, childcare facilities, power plants, forests, dams, fishing ports, stations, airports, and the like, and may be areas under construction or after completion of construction work, and large-scale outdoor facilities that do not involve buildings, such as fields, parks, grounds, and parking lots, may also be monitored. Furthermore, the interiors of moving objects such as buses, trains, ships, airplanes, submarines, and spacecraft may also be monitored.

[0082] The measurement data acquisition unit 13 acquires measurement data for each of a plurality of environmental indicators of the area to be monitored, the plurality of indicators being acquired by a plurality of types of sensors. The measurement data acquiring unit 13 may acquire measurement data directly from the multiple types of sensors, or may acquire measurement data of the multiple types of sensors that is temporarily stored in a server on the Internet 11 from the server. This server refers to, for example, an IoT server, and may be an on-premise server or a cloud server. The multiple indicators refer to odor, air pressure, temperature, humidity, air pollution, smoke concentration, carbon monoxide concentration, carbon dioxide concentration, illuminance, noise level, etc. in the area to be monitored. The measurement data acquisition unit 13 stores the acquired measurement data from the multiple types of sensors in the storage unit 10d.

[0083] The measurement data acquiring unit 13 may be configured such that one or more thresholds are set for the measurement data of each of the multiple indices, and the measurement data acquiring unit 13 acquires measurement data for each of the indices that exceeds the threshold from each of the multiple types of sensors.

[0084] In the environmental measurement terminal 20, one or more thresholds are set for each of the measurement data of the multiple types of sensors, and the CPU 20e transmits the measurement data that it determines to have exceeded the threshold to the fire sign detection system 10. The measurement data acquisition unit 13 acquires the measurement data transmitted from the environmental measurement terminal 20 to the fire sign detection system 10. Since it is expected that the measurement data from multiple types of sensors will be enormous in volume, transmitting only the measurement data that is deemed useful in the fire precursor detection system 10 can reduce the amount of measurement data transmitted from the environmental measurement terminal 20 to the fire precursor detection system 10.

[0085] A single threshold is set when specifying a lower or upper limit for measurement data to be transmitted to the fire sign detection system 10. For example, when a lower threshold is set, measurement data equal to or greater than the set threshold is transmitted from the environmental measurement terminal 20 to the fire sign detection system 10. When an upper threshold is set, measurement data equal to or less than the set threshold is transmitted from the environmental measurement terminal 20 to the fire sign detection system 10.

[0086] Multiple thresholds may be set for one type of measurement data, and two thresholds are set when specifying lower and upper limits for the measurement data to be transmitted to the fire warning detection system 10 and transmitting only measurement data of one band from the environmental measurement terminal 20 to the fire warning detection system 10. For example, when a lower threshold and an upper threshold are set, measurement data having values ​​between the set lower threshold and upper threshold are transmitted from the environmental measurement terminal 20 to the fire warning detection system 10. When measurement data of two bands is transmitted from the environmental measurement terminal 20 to the fire warning detection system 10, a total of four thresholds are specified for the measurement data of one index.

[0087] The sign presence / absence determining unit 14 determines the presence or absence of a sign of a fire in the area to be monitored based on a combination of measurement data of predetermined indices selected from the measurement data acquired by the measurement data acquiring unit 13. The sign presence / absence determination unit 14 refers to a table that defines the correspondence between the type of area to be monitored and combinations of measurement data of environmental indicators of the area, and determines a combination of measurement data of indicators based on the type of area to be monitored accepted by the accepting unit 12.

[0088] The table will be described with reference to Fig. 8. Fig. 8 is a diagram showing an example of a table that defines the correspondence between the type of area to be monitored and a combination of measurement data of a plurality of indexes of the environment of the area. The table specifies, for each type of area to be monitored, a combination of measurement data of indices that the sign presence / absence determination unit 14 uses to determine the presence or absence of a fire sign in the area to be monitored. The combinations specified in the table are merely examples, are not limited to these, and may be changed.

[0089] When the area to be monitored is a construction site where a building is being constructed, the warning sign determination unit 14 determines whether there are warning signs of a fire in the area to be monitored based on a combination of measurement data of odor, temperature, carbon dioxide concentration, and oxygen concentration. When the area to be monitored is a construction site where a logistics warehouse is being built, the warning sign determination unit 14 determines whether there are warning signs of a fire in the area to be monitored based on a combination of measurement data of odor, temperature, carbon dioxide concentration, and oxygen concentration.

[0090] When the area to be monitored is a construction site of a petrochemical plant, the warning signs determination unit 14 determines whether there are warning signs of a fire in the area to be monitored based on a combination of measurement data of odor, temperature, air pollution, carbon monoxide concentration, and oxygen concentration. When the area to be monitored is a construction site of a machinery factory under construction, the warning sign determination unit 14 determines whether there are warning signs of a fire in the area to be monitored based on a combination of measurement data of odor, temperature, smoke concentration, carbon dioxide concentration, and oxygen concentration.

[0091] When the area to be monitored is a construction site of a metal factory under construction, the warning sign determination unit 14 determines whether there are warning signs of a fire in the area to be monitored based on a combination of measurement data of odor, temperature, smoke concentration, carbon dioxide concentration, and oxygen concentration. When the area to be monitored is a construction site of a food factory, the warning signs determination unit 14 determines whether there are warning signs of a fire in the area to be monitored based on a combination of measurement data of odor, humidity, carbon monoxide concentration, and oxygen concentration.

[0092] When the area to be monitored is a construction site of a pharmaceutical factory, the warning sign determination unit 14 determines whether there are warning signs of a fire in the area to be monitored based on a combination of measurement data of odor, air pressure, carbon monoxide concentration, carbon dioxide concentration, and oxygen concentration. When the area to be monitored is a construction site of a paper factory under construction, the warning sign determination unit 14 determines whether there are warning signs of a fire in the area to be monitored based on a combination of measurement data of odor, air pressure, temperature, carbon dioxide concentration, and oxygen concentration.

[0093] When the area to be monitored is a construction site of a semiconductor factory, the warning sign determination unit 14 determines whether there are warning signs of a fire in the area to be monitored based on a combination of measurement data of odor, temperature, carbon monoxide concentration, carbon dioxide concentration, and oxygen concentration.

[0094] The determination of the presence or absence of a sign of a fire in the area to be monitored by the sign presence determination unit 14 will be described with reference to Fig. 9. Fig. 9 is a diagram for explaining an example of a combination of measurement data of a plurality of indexes of the environment of the area to be monitored.

[0095] In the example shown in FIG. 9, the sign presence / absence determination unit 14 determines the presence or absence of signs of a fire in the area to be monitored based on a combination of measurement data of carbon monoxide concentration, temperature, and carbon dioxide concentration as multiple indices of the environment of the area to be monitored. A graph line 41 of carbon monoxide concentration, a graph line 42 of temperature, and a graph line 43 of carbon dioxide concentration each show a change over time. The sign presence / absence determination unit 14 determines the presence or absence of a sign of a fire in the area to be monitored by looking at the manner in which the measurement data of carbon monoxide concentration, temperature, and carbon dioxide concentration change over time from a composite perspective.

[0096] Therefore, the presence or absence of a fire sign in the area to be monitored is determined based on the correlation of the measurement data of carbon monoxide concentration, temperature, and carbon dioxide concentration without setting a threshold value for each of the measurement data. Correlation refers to a relationship in which two or more things are closely related and when one changes, the others also change.

[0097] When the sign presence / absence determining unit 14 determines that there is a sign of a fire in the area to be monitored, the notification unit 16 notifies a person monitoring the area that there is a sign of a fire. The notification unit 16 notifies the observer of the area by email, push notification, calling a specified telephone number, sounding an alarm, etc. A push notification is a notification that is made using the functions of an application on a smartphone or the like.

[0098] The warning sign determination unit 14 uses an warning sign learning model that has pre-learned the correspondence between combinations of measurement data of specified indicators of the environment of the monitored area and the presence or absence of warning signs of a fire in the area, and determines the presence or absence of warning signs of a fire in the monitored area by inputting combinations of measurement data of specified indicators selected from the measurement data acquired by the measurement data acquisition unit 13 into the warning sign learning model.

[0099] The supervised learning data for the predictive learning model is created as follows. Learning data annotated with fire warning signs is generated as a set of data by adding metadata to a combination of measurement data of environmental indicators of an area being monitored for conditions that indicate fire warning signs in that area.

[0100] Learning data annotated with "no signs of fire" is generated as a set of data by adding metadata to a combination of measurement data of environmental indicators of a monitored area when there are no signs of fire, indicating that the combination of measurement data showed no signs of fire in that area. By creating a large amount of training data that is annotated with indications of fire and a large amount of training data that is annotated with indications of no fire, and by machine learning the predictive learning model using the large amount of training data, the predictive accuracy of the predictive learning model can be improved. In addition, learning data for the predictive learning model needs to be created for each type of area to be monitored.

[0101] Furthermore, the learning data for the predictive learning model is newly created using measurement data of the environment of the area to be monitored, obtained by operating the fire predictive detection system 10. Every time a predetermined number of such measurement data is accumulated, new learning data is created and machine learning is applied to the predictive learning model, thereby increasing the amount of learning and improving the prediction accuracy of the predictive learning model.

[0102] When the warning sign determination unit 14 determines that there are signs of a fire in the area to be monitored, the danger level determination unit 15 determines the danger level of the signs of a fire based on a combination of measurement data of multiple indicators of the environment of the specified area that was the basis for the determination by the warning sign determination unit 14. When the warning sign determination unit 14 determines that there are signs of a fire in the area being monitored, the alarm unit 16 notifies the person monitoring the area that there are signs of a fire and the level of danger of the signs determined by the danger level determination unit 15.

[0103] The risk level refers to the fire damage that is expected to occur from the detected fire warning signs. Fire damage refers to the combined damage caused by burning and fire extinguishing damage, where burning damage refers to damage caused by burning, destruction, or deterioration due to the flames or high temperatures of a fire, and fire extinguishing damage refers to damage caused by water damage caused during fire extinguishing, and damage caused by breakage or staining during fire extinguishing. Note that the risk level is not limited to this, and other damages may also be considered. For example, the risk level may include the predicted cost of extinguishing the fire damage, or personal damage that is damage related to human life. The risk level (grade) may be divided into multiple categories, for example, 2 categories, 3 categories, or 5 categories. The risk level in this embodiment is divided into three levels, Level 1, Level 2, and Level 3, in descending order of damage.

[0104] The risk assessment unit 15 uses a risk learning model that has previously learned the correspondence between combinations of measurement data of specified indicators of the environment of the area to be monitored and the risk of fire signs in that area, and judges the risk of fire signs in that area by inputting into the risk learning model the combinations of measurement data of the specified indicators that formed the basis for the judgment by the sign presence / absence determination unit 14 that there are signs of a fire in that area.

[0105] The supervised learning data for the risk learning model is created as follows. Training data annotated with a risk level of Level 1 is generated as a set of data by adding metadata to a combination of measurement data of environmental indicators of an area being monitored where the risk level is Level 1, stating that the risk level of fire precursors in that area for that combination of measurement data is Level 1. Training data annotated with a risk level of Level 2 is generated as a set of data by adding metadata to a combination of measurement data of environmental indicators of an area being monitored for a risk level of Level 2, stating that the risk level of fire precursors in that area for that combination of measurement data is Level 2. Training data annotated with a risk level of Level 3 is generated as a set of data by adding metadata to a combination of measurement data of environmental indicators of an area being monitored for a risk level of Level 3, stating that the risk level of fire precursors in that area for that combination of measurement data is Level 3.

[0106] By creating a large amount of learning data annotated with risk levels of level 1, level 2, and level 3, and machine-learning the risk learning model using a large amount of learning data, the predictive accuracy of the risk learning model can be improved. In addition, like the learning data for the predictive learning model, the learning data for the risk learning model needs to be created for each type of area to be monitored.

[0107] Similarly to the learning data of the precursor learning model, the learning data of the risk learning model is newly created using measurement data of the environment of the area to be monitored acquired by operation of the fire precursor detection system 10. Each time a predetermined number of the measurement data is accumulated, new learning data is created and machine learning is applied to the risk learning model, thereby increasing the amount of learning and improving the prediction accuracy of the risk learning model.

[0108] The abnormal value extraction unit 17 performs unsupervised clustering on the measurement data acquired by the measurement data acquisition unit 13, and extracts outliers of the measurement data for each of multiple indicators of the environment of the area to be monitored contained in the measurement data as abnormal values ​​of the measurement data for each of the multiple indicators. The outlier extractor 17 treats the measurement data acquired by the measurement data acquirer 13 as unsupervised learning data, and performs clustering on this unsupervised learning data. An outlier is a value that is extremely small or extremely large compared to others. The abnormal value extraction unit 17 uses an algorithm such as K-means or hierarchical clustering to create groups (clusters) within the measurement data based on the internal similarity of each of the measurement data of multiple indicators of the environment of the monitored area acquired by the measurement data acquisition unit 13, and extracts outliers that do not belong to any of the groups as abnormal values. This clustering is performed on each of the measurement data of multiple indicators of the environment of the monitored area, and abnormal values ​​are extracted for each of the measurement data of the multiple indicators. Clustering is a type of unsupervised learning that targets data without labels (annotations), calculates the similarity (distance) between data, and classifies the data into several groups (clusters). Clustering calculates the distance between all data. Clustering groups data, making it easier to find patterns or structures in the data, and easier to find outliers that fall outside of the group. Therefore, clustering is useful for extracting abnormal values ​​contained in the data. There are two types of clustering algorithms: agglomerative and divisive. Agglomerative algorithms treat each data set as a single cluster at first, and then gradually integrate the clusters. Divisive algorithms treat all data sets as a single cluster at first, and then divide the data set into a number of clusters (groups) that is specified in advance. By using clustering, a type of unsupervised learning, the outlier extraction unit 17 can extract outliers from the vast amount of measurement data acquired by the measurement data acquisition unit 13 without the time-consuming task of creating supervised learning data. The unsupervised learning performed by the fire warning detection system 10 is not limited to clustering, and other unsupervised learning may be performed. The fire warning detection system 10 may use unsupervised learning, for example, to analyze the structure and characteristics of the acquired data itself, to discover frequent patterns in the acquired data, to group the acquired data, or to simplify the acquired data. In addition, the fire warning detection system 10 may use, for example, a generative adversarial network, which is a type of unsupervised learning, and by learning features from the acquired data, it becomes possible to generate non-existent data or convert data in accordance with the features of the acquired data. Furthermore, the fire warning detection system 10 may use, for example, association analysis, which is a type of unsupervised learning, making it possible to discover associations between acquired data. In addition, the fire warning detection system 10 may use, for example, Principal Component Analysis (PCA), which is a type of unsupervised learning, to summarize (reduce dimensions) multiple explanatory variables contained in the acquired data to find new explanatory variables, and make the new explanatory variables the principal components of the data, making it possible to make the data easier to understand.

[0109] Abnormal value sign determination section 18 determines the presence or absence of a sign of a fire in the area based on the abnormal value extracted by abnormal value extraction section 17. Abnormal value sign determination unit 18 may determine that there is a sign of a fire in the area to be monitored when the abnormal value extracted by abnormal value extraction unit 17 exceeds a preset threshold. This threshold is set for each of a plurality of indicators of the environment of the area to be monitored, and abnormal value sign determination unit 18 may determine that there is a sign of a fire in the area to be monitored when an abnormal value in any of the measurement data of each of the plurality of indicators exceeds the set threshold, or may determine that there is a sign of a fire in the area to be monitored when abnormal values ​​in a predetermined number of indicators exceed the threshold. Alternatively, the abnormal value precursor determination unit 18 may use a learning model that has previously learned the correspondence between the abnormal values ​​extracted by the abnormal value extraction unit 17 and the presence or absence of fire precursors in the area to be monitored, to determine the presence or absence of fire precursors in the area to be monitored based on the abnormal values ​​extracted by the abnormal value extraction unit 17. Abnormal value sign determination unit 18 determines the presence or absence of signs of a fire in the area based on the abnormal values ​​extracted by abnormal value extraction unit 17, and sign presence determination unit 14 determines the presence or absence of signs of a fire in the area to be monitored based on a combination of measurement data of predetermined indices selected from the measurement data acquired by measurement data acquisition unit 13. Therefore, abnormal value sign determination unit 18 and sign presence determination unit 14 can determine the presence or absence of signs of a fire in the area to be monitored based on two different criteria (one is an abnormal value, the other is a combination of measurement data).

[0110] The threshold resetting unit 19 resets the threshold for the measurement data of each of the multiple indexes based on the abnormal value extracted by the abnormal value extracting unit 17. The threshold resetting unit 19 resets the thresholds used by the measurement data acquisition unit 13 when acquiring measurement data from each of multiple types of sensors, based on the abnormal values ​​extracted by the abnormal value extraction unit 17 for each of multiple indicators of the environment of the area to be monitored. The threshold resetting unit 19 may set the threshold to, for example, 70% of the abnormal value, but the threshold is not limited to 70% and may be any value between 1% and 99%. The measurement data acquiring unit 13 acquires measurement data that exceeds the threshold value reset by the threshold value resetting unit 19 for each of the multiple indices from each of the multiple types of sensors. The threshold resetting unit 19 may set multiple thresholds for one type of measurement data, and resets the lower and upper limits of the measurement data to be transmitted to the fire sign detection system 10, and resets two thresholds when only measurement data of one band is transmitted from the environmental measurement terminal 20 to the fire sign detection system 10. For example, when the lower and upper thresholds are reset, measurement data having values ​​ranging from the reset lower threshold to the reset upper threshold is transmitted from the environmental measurement terminal 20 to the fire sign detection system 10. When measurement data of two bands is transmitted from the environmental measurement terminal 20 to the fire sign detection system 10, a total of four thresholds are reset for the measurement data of one index. Threshold resetting section 19 can reset the threshold used by measurement data acquiring section 13 based on the abnormal value extracted by abnormal value extracting section 17.

[0111] (Experimental results of Fire Prediction Detection System 10) Next, experimental results of the fire sign detection system 10 will be described with reference to Fig. 10. Fig. 10 is a diagram for explaining an example of experimental results of the fire sign detection system 10.

[0112] Fig. 10 shows the measurement results of the temperature distribution of the two-dimensional thermopile radiation temperature sensor 22, which divides the two-dimensional plane of the area to be monitored into a total of 64 sections arranged in an 8 x 8 matrix. Each square in Fig. 10 corresponds to one section of the two-dimensional plane, and the numbers in each square indicate the temperature (°C) of that section. As shown in Fig. 10, the temperature distribution measurement results of the two-dimensional thermopile radiation temperature sensor 22 indicate the section with the highest temperature, thereby suggesting the section in question that may be the cause of a fire. Furthermore, by changing the color of the squares indicating the temperature distribution measurement results according to the temperature of the section, the temperature distribution can be easily recognized visually.

[0113] (About the fire warning method and the fire warning program) Next, a fire sign detection program according to one embodiment of the present invention will be described together with a fire sign detection method with reference to Fig. 11. Fig. 11 is an example of a flowchart of the fire sign detection program according to this embodiment.

[0114] The fire sign detection method is executed by the CPU 10e of the fire sign detection system 10 based on a fire sign detection program. As shown in FIG. 11, the fire sign detection program includes a receiving step S12, a measurement data acquiring step S13, a sign presence / absence determining step S14, a danger level determining step S15, and a notification step S16.

[0115] The fire sign detection program causes the CPU 10e of the fire sign detection system 10 to realize various functions, including a reception function, a measurement data acquisition function, a sign presence / absence determination function, a danger level determination function, and a notification function. These functions are executed in the order shown in the flowchart of Fig. 11, but the order can be changed as appropriate. Note that each function overlaps with the description of the various functional parts of the fire sign detection system 10 described above, and therefore detailed description thereof will be omitted.

[0116] The reception function receives the type of area to be monitored (S12: reception step). The reception function receives the type of area to be monitored, which is input by a user of the fire warning detection system 10 or an owner or supervisor of the area to be monitored.

[0117] The measurement data acquisition function acquires measurement data of each of a plurality of environmental indicators of the area to be monitored, which are acquired by a plurality of types of sensors (S13: measurement data acquisition step). The measurement data acquisition function acquires measurement data from a plurality of types of sensors from the environment measurement terminal 20.

[0118] The warning signs determination function determines whether or not there are warning signs of a fire in the area being monitored based on a combination of measurement data of specified indicators selected from the measurement data acquired by the measurement data acquisition function (S14: warning signs determination step).

[0119] When the warning sign determination function determines that there are signs of a fire in the area being monitored, the danger assessment function determines the danger level of the warning signs based on a combination of measurement data of multiple indicators of the environment of the specified area that served as the basis for the determination made by the warning sign determination function (S15: danger determination step).

[0120] When the sign presence / absence determination function determines that there is a sign of a fire in the area to be monitored, the notification function notifies a person monitoring the area that there is a sign of a fire (S16: notification step).

[0121] (Fire sign detection method and fire sign detection program according to other embodiments) A fire sign detection program according to another embodiment will be described together with a fire sign detection method with reference to Fig. 12. Fig. 12 is an example of a flowchart of a fire sign detection program according to another embodiment. The flowchart of a fire warning detection program according to another embodiment shown in FIG. 12 differs from the flowchart of a fire warning detection program shown in FIG. 11 in that an abnormal value extraction step S17, an abnormal value warning determination step S18, and a threshold resetting step S19 are added. A fire sign detection method according to another embodiment is executed by the CPU 10e of the fire sign detection system 10 based on a fire sign detection program according to another embodiment shown in FIG. The fire warning detection program according to another embodiment shown in FIG. 12 includes a reception step S12, a measurement data acquisition step S13, a warning presence / absence determination step S14, a danger level determination step S15, an alarm step S16, an abnormal value extraction step S17, an abnormal value warning presence determination step S18, and a threshold resetting step S19.

[0122] 12 causes the CPU 10e of the fire sign detection system 10 to realize a reception function, a measurement data acquisition function, a sign presence / absence determination function, a risk level determination function, a notification function, an abnormal value extraction function, an abnormal value sign determination function, a threshold resetting function, etc. These functions are executed in the order shown in the flowchart of FIG. 12, but the order can also be changed as appropriate. Hereinafter, the fire sign detection method and fire sign detection program according to another embodiment shown in FIG. 12 will be described, focusing only on the differences from the fire sign detection method and fire sign detection program shown in FIG. Furthermore, since each function overlaps with the description of the various functional parts of the fire warning detection system 10 described above, detailed description thereof will be omitted.

[0123] The outlier extraction function performs unsupervised clustering of the measurement data acquired by the measurement data acquisition function to extract outliers in the measurement data for each of multiple indicators of the environment of the area included in the measurement data as outliers in the measurement data for each of the multiple indicators (S17: outlier extraction step).

[0124] The abnormal value sign determination function determines whether or not there is a sign of a fire in the area based on the abnormal value (S18: abnormal value sign determination step).

[0125] The threshold resetting function resets the threshold for the measurement data of each of the multiple indexes based on the abnormal value (S19: threshold resetting step).

[0126] According to the embodiment of the fire warning detection system 10 described above, the area to be monitored is monitored using multiple sensors, including an odor sensor 21, a two-dimensional thermopile radiation temperature sensor 22, an air pressure sensor 23, a temperature and humidity sensor 24, an air pollution sensor 25, a smoke sensor 26, a carbon monoxide sensor 27, a carbon dioxide sensor 28, an illuminance sensor 29, a noise sensor 30, and a camera with night vision function 31, to monitor the area using multiple environmental indicators, thereby making it possible to detect warning signs of a fire early.

[0127] Furthermore, according to the fire warning detection system 10, the presence or absence of warning signs of a fire in a monitored area is determined using a machine-learned warning learning model based on a combination of measurement data of multiple environmental indicators in the area. Therefore, compared to a case in which a threshold is set for the measurement data of the environmental indicators and the presence or absence of warning signs of a fire is determined based on whether the measurement data exceeds the threshold, the presence or absence of warning signs of a fire is determined by taking into account the correlation between multiple environmental indicators without fixing a threshold for the measurement data, so that a more realistic judgment can be made earlier.

[0128] Furthermore, according to the fire warning detection system 10, a combination of measurement data of environmental indicators of the area to be monitored is determined according to the type of area, so that the presence or absence of warning signs of a fire can be determined according to the characteristics of the area, thereby making it possible to more quickly determine the presence or absence of warning signs of a fire.

[0129] Furthermore, according to the fire sign detection system 10, the risk of an area determined to have a fire sign is determined using a risk learning model that has been machine-learned, thereby making it possible to determine the risk more accurately.

[0130] Furthermore, the fire sign detection system 10 can determine the presence or absence of a sign of a fire in a monitored area by taking into account the oxygen concentration in the area.

[0131] Furthermore, the fire sign detection system 10 can determine the risk of a fire sign by taking into account the oxygen concentration in the area to be monitored.

[0132] Furthermore, according to the fire warning detection system 10, the unmanned aerial vehicle 35 or robot 36 is equipped with an environmental measurement terminal 20, allowing the user to move the environmental measurement terminal 20 to an area of ​​their choice and determine whether or not there are warning signs of a fire in that area.

[0133] Furthermore, according to the fire warning detection system 10, by using clustering, a type of unsupervised learning, it is possible to extract abnormal values ​​from the vast amount of measurement data acquired by the measurement data acquisition unit 13, without the time-consuming task of creating supervised learning data.

[0134] Furthermore, according to the fire warning detection system 10, the abnormal value warning determination unit 18 and the warning presence / absence determination unit 14 can determine the presence or absence of warning signs of a fire in the area being monitored based on two different criteria (one being an abnormal value and the other being a combination of measurement data).

[0135] Furthermore, according to the fire warning detection system 10, the threshold resetting unit 19 can reset the threshold used by the measurement data acquiring unit 13 based on the abnormal value extracted by the abnormal value extracting unit 17.

[0136] The present invention is not limited to the fire warning detection system 10, the fire warning detection method, and the fire warning detection program according to the above-described embodiments, but can be implemented by various other modified examples or application examples without departing from the gist of the present invention as described in the claims. [Explanation of symbols]

[0137] 10. Fire Prediction and Detection System 10a Communication Interface 10b ROM 10c RAM 10d storage section 10e CPU 10f Input / Output Interface 10g Input Device 10h output device 11. Internet 12 Reception 13 Measurement data acquisition section 14 Prediction Unit 15 Risk Assessment Section 16. Notification Department 17 Outlier extraction section 18 Abnormal value prediction determination unit 19 Threshold resetting unit 20 Environmental measurement terminal 20a Communication Interface 20b ROM 20c RAM 20d storage section 20e CPU 20f Input / Output Interface 21 Odor Sensor 22 Two-dimensional thermopile radiation temperature sensor 23 Barometric pressure sensor 24 Temperature and humidity sensor 25 Air pollution sensor 26 Smoke Sensor 27 Carbon monoxide sensor 28 Carbon Dioxide Sensor 29 Illuminance sensor 30 Noise Sensor 31 Night Vision Camera 32 Oxygen sensor 33 Ceiling 34 Stand 35 Unmanned Aerial Vehicles 36 Four-legged robot 37 Helmet 41 Carbon monoxide concentration graph line 42 Temperature graph line 43 Carbon dioxide concentration graph line

Claims

1. A fire warning detection system that is capable of communicating with a plurality of types of sensors that are installed in a monitored area and measure the environment of the area, and detects warning signs of a fire in the area, a measurement data acquisition unit that acquires measurement data of each of a plurality of environmental indicators of the area acquired by the plurality of types of sensors; a sign presence / absence determination unit that determines the presence or absence of a sign of a fire in the area based on a combination of measurement data of predetermined indicators selected from the measurement data acquired by the measurement data acquisition unit; a notification unit that notifies a monitor of the area that there is a sign of a fire when the sign determination unit determines that there is a sign of a fire in the area; A fire warning system comprising:

2. The fire warning detection system according to claim 1, characterized in that the warning sign determination unit uses a warning sign learning model that has previously learned the correspondence between combinations of measurement data of predetermined indicators and the presence or absence of warning signs of a fire in the area, and determines the presence or absence of warning signs of a fire in the area by inputting into the warning sign learning model combinations of measurement data of predetermined indicators selected from the measurement data acquired by the measurement data acquisition unit.

3. A reception unit that receives a type of the area to be monitored, The sign presence / absence determination unit is The fire warning system according to claim 1, characterized in that the combination of measurement data of the indicators is determined based on the type of area to be monitored received by the receiving unit, by referring to a table that defines a correspondence between the type of area to be monitored and the combination of measurement data of the indicators.

4. a danger level determination unit that, when the sign presence / absence determination unit determines that there is a fire sign in the area, determines a danger level of the fire sign based on a combination of measurement data of the predetermined indicators that are the basis for the determination by the sign presence / absence determination unit, The fire warning system according to claim 1, characterized in that, when the warning presence / absence determination unit determines that there are signs of a fire in the area, the alarm unit notifies a person monitoring the area that there are signs of a fire and the level of danger of the signs determined by the danger determination unit.

5. The fire warning detection system of claim 4, wherein the danger assessment unit uses a danger learning model that has previously learned the correspondence between combinations of measurement data of predetermined indicators and the danger of fire warning signs in the area, and judges the danger of fire warning signs in the area by inputting into the danger learning model the combinations of measurement data of predetermined indicators that are the basis for the warning sign presence / absence determination unit's determination that there are warning signs of a fire in the area.

6. The measurement data acquisition unit One or more thresholds are set for the measurement data of each of the plurality of indices, and the measurement data exceeding the threshold for each of the indices is acquired from each of the plurality of types of sensors.

2. The fire warning system according to claim 1,

7. 2. The fire warning detection system according to claim 1, wherein the plurality of types of sensors includes an odor sensor that measures odors in the area.

8. The fire warning system according to claim 1, characterized in that the multiple types of sensors include a two-dimensional thermopile radiation temperature sensor that divides a two-dimensional plane of the area into predetermined sections and measures a two-dimensional temperature distribution that indicates the temperature of each section as a numerical value.

9. 2. The fire warning system according to claim 1, wherein the plurality of types of sensors includes a barometric pressure sensor for measuring barometric pressure in the area.

10. 2. The fire warning system according to claim 1, wherein the plurality of types of sensors includes a temperature and humidity sensor that measures temperature and humidity in a steady state in the area.

11. 2. The fire warning system according to claim 1, wherein the plurality of types of sensors includes an air pollution sensor that measures air pollution in the area.

12. 2. The fire warning system according to claim 1, wherein the plurality of types of sensors includes a smoke sensor that detects a concentration of smoke in the area.

13. 2. The fire warning system according to claim 1, wherein the plurality of types of sensors includes a carbon monoxide sensor that detects a concentration of carbon monoxide in the area.

14. 2. The fire warning system according to claim 1, wherein the plurality of types of sensors includes a carbon dioxide sensor that detects a concentration of carbon dioxide in the area.

15. 2. The fire warning system according to claim 1, wherein the plurality of types of sensors includes an illuminance sensor for measuring an illuminance of the area.

16. 2. The fire warning system according to claim 1, wherein the plurality of types of sensors includes a noise sensor that measures the volume of noise in the area.

17. 2. The fire warning system according to claim 1, wherein the plurality of types of sensors includes an oxygen sensor that measures an oxygen concentration in the area.

18. The plurality of types of sensors are mounted on an unmanned aerial vehicle or a robot, The fire warning system according to claim 1, characterized in that the unmanned aerial vehicle or robot flies or runs in the area by remote control or autonomous control.

19. an abnormal value extraction unit that performs unsupervised clustering of the measurement data acquired by the measurement data acquisition unit to extract outliers of the measurement data for each of a plurality of indexes of the environment of the area included in the measurement data as abnormal values ​​of the measurement data for each of the plurality of indexes; an abnormal value sign determination unit that determines whether or not there is a sign of a fire in the area based on the abnormal value; The fire warning system according to claim 1, further comprising:

20. an abnormal value extraction unit that performs unsupervised clustering of the measurement data acquired by the measurement data acquisition unit to extract outliers of the measurement data for each of a plurality of indexes of the environment of the area included in the measurement data as abnormal values ​​of the measurement data for each of the plurality of indexes; and a threshold resetting unit that resets the threshold for each of the measurement data of the plurality of indexes based on the abnormal value, The measurement data acquisition unit The measurement data exceeding the threshold reset by the threshold resetting unit for each of the plurality of indices is acquired from each of the plurality of types of sensors.

7. A fire warning system according to claim 6.

21. A fire warning detection method for detecting a warning sign of a fire in a monitored area, the method being capable of communicating with a plurality of types of sensors that are installed in the monitored area and measure the environment of the monitored area, the method comprising: The computer a measurement data acquisition step of acquiring measurement data of each of a plurality of indicators of the environment of the area acquired by the plurality of types of sensors; a sign presence / absence determination step for determining whether or not there is a sign of a fire in the area based on a combination of measurement data of predetermined indicators selected from the measurement data acquired in the measurement data acquisition step; a notification step of notifying a supervisor of the area that there is a fire sign when it is determined that there is a fire sign in the area in the sign presence / absence determination step; A fire warning detection method comprising the steps of:

22. A fire warning detection program that is capable of communicating with a plurality of types of sensors that are installed in a monitored area and measure the environment of the area, and detects warning signs of a fire in the area, On the computer, a measurement data acquisition function for acquiring measurement data of each of a plurality of environmental indicators of the area acquired by the plurality of types of sensors; a sign presence / absence determination function that determines the presence or absence of a sign of a fire in the area based on a combination of measurement data of predetermined indicators selected from the measurement data acquired by the measurement data acquisition function; a warning function for warning a person watching the area that there is a fire warning when the warning function determines that there is a fire warning in the area; and A fire warning detection program characterized by:

Images