JP7357457B2 - Fire alarm system, information processing device, fire alarm method and program - Google Patents

Description

The present invention relates to a fire alarm system, an information processing device, a fire alarm method, and a program.

Conventionally, in general houses, fire alarms have been installed in kitchens, bedrooms, stairs, etc. As such fire alarms, those using fire detection sensors such as smoke sensors and heat sensors are generally used. Furthermore, conventionally, a system has been proposed that determines the presence or absence of a fire in the vicinity of the installation position of the fire detection sensor based on the sensing results of the fire detection sensor, and provides a notification when a fire occurs (for example, Patent Document 1 reference).

JP2018-60578A

By the way, there is a non-fire alarm problem with the above-mentioned fire alarm. Specifically, fire alarms use fire detection sensors that detect fire using smoke and heat as indicators, so non-fire alarms are generated by reacting to smoke and heat other than fire. Although it is possible to reduce non-fire alarms by adjusting the sensitivity of the fire detection sensor, there is a possibility that actual signs of fire cannot be detected, and there is a trade-off relationship with the occurrence of non-fire alarms. Further, even in the conventional system described above, since a fire detection sensor that detects smoke and heat is used, it is difficult to fundamentally reduce the occurrence of non-fire alarms.

The present invention has been made in view of the above, and provides a fire alarm system, an information processing device, a fire alarm method, and a program that can achieve both early detection of fire signs and a reduction in the occurrence of non-fire alarms. The purpose is to

In order to solve the above-mentioned problems and achieve the objectives, a fire alarm system according to the present invention is a fire alarm system that includes a sensor unit and an information processing device connected to the sensor unit via a network. , the sensor unit includes an odor sensor that detects odor, a motion sensor that detects human body movement, and a fire detection sensor that detects smoke or heat, and the information processing device detects the information from the sensor unit. an acquisition unit that acquires sensing results around the installation location where the sensor unit is installed; and a first determination unit that determines whether or not there is a suspicion of a fire based on the sensing results of the odor sensor acquired by the acquisition unit. and, when the first determination unit determines that a fire is suspected, the acquisition unit determines the occurrence of a fire around the installation location based on the sensing results of the human sensor or the fire detection sensor. The information processing apparatus includes a second determination section that makes a determination, and a reporting section that reports to a preset reporting destination according to the determination result of the second determination section.

According to the present invention, it is possible to achieve both early detection of fire signs and a reduction in the occurrence of non-fire alarms.

FIG. 1 is a diagram showing an example of the configuration of a fire alarm system according to an embodiment. FIG. 2 is a diagram illustrating an example of the hardware configuration of the cloud service device according to the embodiment. FIG. 3 is a diagram showing an example of the data structure of the sensor management file according to the embodiment. FIG. 4 is a diagram illustrating an example of the data structure of the reporting destination management file according to the embodiment. FIG. 5 is a diagram schematically showing the operation of the situation determination data (trained model) according to the embodiment. FIG. 6 is a diagram illustrating an example of the functional configuration of the cloud service device according to the embodiment. FIG. 7 is a sequence diagram showing an example of the operation of the fire alarm system according to the embodiment. FIG. 8 is a flowchart illustrating an example of processing executed by the cloud service device according to the embodiment. FIG. 9 is a diagram schematically showing the operation of the situation determination data (learned model) according to the modified example.

Hereinafter, a fire alarm system, an information processing device, a fire alarm method, and a program according to embodiments will be described with reference to the drawings. In addition, although the embodiment described below describes an example applied to a general house, the present invention is not limited to the following embodiment.

FIG. 1 is a diagram showing an example of the configuration of a fire alarm system according to an embodiment. As shown in FIG. 1, the fire alarm system 1 includes a sensor unit 10 and a cloud service device 20.

The sensor unit 10 is installed in an area within the house H to be monitored. For example, the sensor units 10 are installed in areas such as a kitchen, a bedroom, and stairs. Note that the sensor unit 10 may be installed at any particular location; for example, it may be installed on the ceiling or the like.

The sensor unit 10 includes multiple types of sensors. Specifically, the sensor unit 10 includes an odor sensor 11, a human sensor 12, and a smoke sensor 13 as a set. That is, the sensing range of the set of odor sensor 11, human sensor 12, and smoke sensor 13 is around the installation position where the sensor unit 10 is installed. Each sensor of the sensor unit 10 may be configured integrally or separately. In addition, when using separate bodies, each sensor may be installed at a different position, but it is preferable to set so that a part or all of the sensing range of each sensor overlaps.

The odor sensor 11 is a sensor device that detects odor. The odor sensor 11 senses the area around the installation position and outputs odor components present around the installation position as a sensing result. The sensing result includes a sensor identifier that allows the odor sensor 11 to be identified.

The configuration of the odor sensor 11 is not particularly limited, and various configurations can be adopted. As an example, the odor sensor 11 may be a QCM (Quartz Crystal Microbalance) sensor in which a film that adsorbs odor components (causative substances) is provided on the surface of a quartz crystal resonator. In a QCM sensor, the resonance frequency of an AT-cut crystal oscillator changes due to a change in mass. The QCM sensor detects the odor of a causative substance as the mass of the causative substance by vibrating an AT-cut crystal resonator and detecting the amount of change in the resonance frequency. Note that the type and number of causative substances to be detected are not particularly limited, but it is possible to detect at least odor-causing substances related to fire, such as odor-causing substances generated when fire spreads in building materials. It shall be possible.

The human sensor 12 is a sensor device that detects the movement of a human body. The human sensor 12 senses the area around the installation position and outputs the amount of movement representing the movement of the human body as a sensing result. The sensing result includes a sensor identifier that can identify the human sensor 12. The configuration of the human sensor 12 is not particularly limited, and various configurations can be adopted. For example, the human sensor 12 may be a human sensor that uses infrared rays, ultrasonic waves, or the like.

The smoke detection sensor 13 is an example of a fire detection sensor, and is a sensor device that detects (senses) smoke.The smoke detection sensor 13 senses the area around the installation location and determines whether smoke is occurring around the installation location. Information indicating this is output as a sensing result. The sensing result includes a sensor identifier that allows the smoke detection sensor 13 to be identified. The configuration of the smoke detection sensor 13 is not particularly limited, and various configurations can be adopted. For example, the smoke sensor 13 may be a photoelectric smoke sensor.

Sensing results from the sensor unit 10 (odor sensor 11, human sensor 12, smoke sensor 13) described above are sent to the cloud service device 20 connected to the network N1 via the access point AP installed in the house H. sent to. Here, the network N1 is, for example, the Internet or a mobile communication network.

Note that in this embodiment, the sensing result of the sensor unit 10 is transmitted to the cloud service device 20 via the access point AP, but the present invention is not limited to this. It is also possible to send the information directly. Further, the sensing results of the odor sensor 11, the human sensor 12, and the smoke sensor 13 may be output at once as the sensing results of the sensor unit 10, or may be output individually.

The cloud service device 20 is an example of an information processing device. The cloud service device 20 is configured by one or more server devices. The cloud service device 20 is connected to one or more houses H to be monitored via a network N1. In FIG. 1, an example is shown in which two houses H (house H1 and house H2) are monitored. The cloud service device 20 receives sensing results from the sensor units 10 installed in each of the houses H, and when detecting signs of a fire or occurrence of a fire from the sensing results, reports (notifies) a predetermined reporting destination.

For example, when the cloud service device 20 is configured as a server device housed in one housing, the cloud service device 20 is realized with a hardware configuration as shown in FIG. 2. Here, FIG. 2 is a diagram showing an example of the hardware configuration of the cloud service device 20.

The cloud service device 20 includes a CPU (Central Processing Unit) 21, a ROM (Read Only Memory) 22, a RAM (Random Access Memory) 23, a storage section 24, and a communication section 25. These units are communicably connected to each other via a bus line.

The CPU 21 is an example of a processor, and controls each part of the cloud service device 20 in an integrated manner by executing programs stored in the ROM 22 and the storage unit 24. The ROM 22 stores programs and the like executed by the CPU 21. The RAM 23 is used by the CPU 21 as a work area for temporarily storing values and the like necessary for executing programs.

The storage unit 24 is a storage device such as an HDD (Hard Disk Drive) or an SSD (Solid State Drive). The storage unit 24 stores programs executable by the CPU 21 and various setting information. The storage unit 24 also stores a sensor management file 241 that defines the relationship between each sensor and its installation position, a reporting destination management file 242 that defines the reporting destination for each house H, and situation determination data related to determining the fire situation. 243.

FIG. 3 is a diagram showing an example of the data structure of the sensor management file 241. As shown in FIG. 3, the sensor management file 241 stores the installation location where each sensor (sensor unit 10) is installed and the sensor identifier of each sensor in association with each other. Here, the installation location of the sensor unit 10 is represented by a combination of a house name, which is an example of an identifier for identifying the house H, and an area within the house H, for example. The sensor identifier is an identifier for identifying each sensor (odor sensor 11, human sensor 12, and smoke sensor 13).

By referring to the sensor management file 241, the cloud service device 20 identifies the installation location and type of the sensor that outputs the sensing result from the sensor identifier included in the sensing result. For example, if the sensor identifier included in the sensing result received from house H is “odor sensor_1011,” the cloud service device 20 determines that the sensor identifier is the sensing result of an odor detected in the “kitchen” of house name “H1.” Identify.

FIG. 4 is a diagram showing an example of the data structure of the report destination management file 242. As shown in FIG. 4, the report destination management file 242 stores the house name of each house H and the report destination in association with each other. Here, the reporting destinations are classified into normal reporting destinations and emergency reporting destinations. The normal reporting destination means the reporting destination when a fire is suspected to occur, such as when a fire sign is detected, and the emergency reporting destination means the reporting destination when the outbreak of a fire is detected. For example, the address (telephone number, e-mail address, etc.) of the owner or resident of the house H, the management company that manages the house H, etc. can be registered as the normal report destination. Further, it is preferable to register the address of an organization capable of dealing with fires, such as a fire station in the area where house H is located, as the emergency call destination.

When the cloud service device 20 determines the occurrence of a fire based on the sensing results, it refers to the reporting destination management file 242 and sends a report to the reporting destination according to the determination result.

The situation determination data 243 is data related to determination of the fire occurrence situation. In this embodiment, the situation determination data 243 is used to determine whether or not there is a suspicion of a fire based on the sensing results of the odor sensor 11.

As an example, the situation determination data 243 is threshold data that defines a threshold value (mass) that serves as a determination index for a specific odor component that occurs during a fire, for example, an odor component that occurs when housing building materials are burned. It's okay. In addition, as another example, the situation determination data 243 includes learning that can infer the possibility of a fire occurring based on probability values (hereinafter referred to as fire probability) from the odor components detected by the odor sensor 11. It may be a completed model.

In the latter case, the trained model can be created, for example, by the following method. First, various odor components are collected using the odor sensor 11 or the like, and each of the sampled odor components is used as learning data. The learning data includes, for example, odor components generated during a fire and odor components collected inside the house H during non-fire times. In addition, data indicating whether or not the odor component originates from a fire as a binary value (1, 0) is prepared as teacher data. Then, by inputting the learning data and teacher data into a creation device that creates a trained model and performing machine learning on the relationship between the learning data and the teacher data, the probability of fire is calculated from the sensing results of the odor sensor 11. It is possible to create a trained model that can be output as an inference result.

FIG. 5 is a diagram schematically showing the operation of the situation determination data 243 (trained model). When the sensing result (odor component) of the odor sensor 11 is input to the trained model created by the above-described creation method, the trained model outputs the fire probability as an inference result. This fire probability is expressed as a value between 0 and 1, for example. For example, in the teacher data, if it is defined that "1" indicates fire origin and "0" indicates non-fire origin, the larger the fire probability value, the more likely fire is suspected.

Returning to FIG. 2, the communication unit 25 is a communication interface for connecting to the network N1. The communication section 25 communicates with the sensor unit 10 under the control of the CPU 21 via the access point AP connected to the network N1.

Further, the cloud service device 20 has a functional configuration shown in FIG. 6. Here, FIG. 6 is a diagram showing an example of the functional configuration of the cloud service device 20.

As shown in FIG. 6, the cloud service device 20 includes a sensing result acquisition section 211, a source identification section 212, a situation determination section 213, and a notification control section 214 as functional configurations.

Part or all of the functional configuration of the cloud service device 20 may be realized by software (program), or may be realized by a hardware circuit (not shown). In one aspect, each part of the functional configuration described above is realized by cooperation between the CPU 21 and a program stored in the storage unit 24. Note that the program may be provided, for example, by being downloaded via a network from another information processing device connected to the network N1, or may be provided by being pre-installed in a portable storage medium or the like. .

The sensing result acquisition unit 211 is an example of an acquisition unit. The sensing result acquisition unit 211 acquires sensing results transmitted from each of the sensor units 10 via the communication unit 25.

Based on the sensor identifier included in the sensing result acquired by the sensing result acquisition unit 211, the transmission source identification unit 212 identifies the installation location of the sensor of the transmission source that transmitted the sensing result. Specifically, the source identification unit 212 refers to the sensor management file 241 and identifies the set of “house name” and “area” corresponding to the sensor identifier included in the sensing result as the installation location of the source sensor. Specify as.

The situation determining unit 213 is an example of a determining unit. The situation determination unit 213 determines the occurrence of a fire around the installation location identified by the transmission source identification unit 212 based on the sensing results acquired by the sensing result acquisition unit 211. Specifically, the situation determination unit 213 uses the sensing results of the odor sensor 11, the human sensor 12, and the smoke sensor 13 to comprehensively determine the fire occurrence situation. An example of the operation of the situation determination unit 213 will be described below.

First, the situation determination unit 213 uses the situation determination data 243 to determine from the sensing result of the odor sensor 11 whether the situation is such that a fire is suspected. Here, the reason why the sensing result of the odor sensor 11 is determined first is that in the initial stage of a fire outbreak, the odor spreads faster than the flame or smoke. Therefore, by comprehensively determining the sensing results of other sensors based on the sensing results of the odor sensor 11, early detection of fire signs can be achieved.

For example, when the situation determination data 243 is threshold data, the situation determination unit 213 compares the sensing result of the odor component generated during a fire detected by the odor sensor 11 with the value of the threshold data. Here, the situation determination data 243 determines that there is a suspicion of fire (fire sign) when the sensing result of the odor sensor 11 is equal to or greater than a threshold value, and determines that there is no suspicion of a fire when it is less than the threshold value.

On the other hand, when the situation determination data 243 is a trained model, the situation determination unit 213 determines whether a fire is suspected based on the inference result (fire probability) obtained by inputting the sensing results of the odor sensor 11 into the learned model. Determine whether or not there is. For example, the situation determination data 243 determines that there is a suspicion of a fire when the fire probability is 40% or more, and determines that there is no suspicion of a fire when the fire probability is less than 40%. Here, the threshold value related to the fire probability is not particularly limited and can be set arbitrarily.

In addition, based on the sensing results of the smell sensor 11 described above, the situation determination unit 213 determines the situation of fire occurrence around the installation location where the smell sensor 11 is installed based on the sensing results of the human sensor 12 or the smoke sensor 13. Determine.

Specifically, the situation determination unit 213 determines whether the degree of activity of human body movement is less than a threshold value, etc., based on the sensing result of the human sensor 12 that senses the same installation location as the odor sensor 11. For example, if the activity level is less than the threshold, the situation determining unit 213 determines that a fire is suspected. Further, if the activity level is equal to or greater than the threshold value, the situation determining unit 213 determines that there is no suspicion of fire.

Further, the situation determining unit 213 determines whether smoke is generated based on the sensing result of the smoke sensor 13 that senses the same installation location as the odor sensor 11. Here, if it is determined that smoke is occurring, the situation determining unit 213 determines that a fire is occurring. Further, when it is determined that no smoke is generated, the situation determination unit 213 maintains the determination that there is a suspicion of fire.

In this manner, the situation determining unit 213 uses the sensing results of the odor sensor 11, the human sensor 12, and the smoke sensor 13 to comprehensively determine the fire occurrence situation. Therefore, compared to the case where the fire occurrence situation is determined using only the smoke detection sensor 13, it is possible to determine the index value related to the fire occurrence from multiple angles, thereby reducing the number of false detections (non-fire alarms). be able to. In addition, the situation determination unit 213 determines the occurrence of a fire from the sensing results of the human sensor 12 and the smoke sensor 13 based on the sensing results of the odor sensor 11, so that early detection of fire signs can be achieved. .

Note that the order of execution of the determination using the sensing results of the human sensor 12 and the smoke sensor 13 is not particularly limited, and the determinations may be exchanged or executed in parallel. In the present embodiment, when smoke is not detected in the sensing result of the smoke sensor 13, the determination using the sensing result of the human sensor 12 is performed.

The notification control unit 214 is an example of a notification unit. Based on the report destination management file 242, the report control unit 214 makes a report to the report destination according to the determination result of the situation determination unit 213.

Specifically, when the situation determination unit 213 determines that a fire is suspected, the notification control unit 214 sends the address of the normal notification destination associated with the installation location (house name) of the sensor unit 10 related to the determination. is read from the report destination management file 242. The notification control unit 214 then reports information including a message notifying that a fire sign has been detected to the read address.

In addition, when the situation determination unit 213 determines that a fire has occurred, the notification control unit 214 sends the address of the emergency notification destination associated with the installation location (house name) of the sensor unit 10 related to the determination to the notification destination management. Read from file 242. The notification control unit 214 then reports information including a message notifying that a fire has occurred to the read address.

Please note that there are no particular restrictions on the method of reporting to the destination. For example, the notification control unit 214 may cooperate with the communication unit 25 to send an e-mail or a message to the address of the notification destination (e-mail address, telephone number, etc.). Further, for example, the notification control unit 214 may use voice synthesis technology or the like to send a voice notification to the address (telephone number) of the destination.

Further, the notification control unit 214 may also notify the installation location of the sensor unit 10, the address of the residence H, etc. when reporting to the notification destination. Further, the notification control unit 214 may also notify the human sensor 12 of information indicating whether or not a person is present in the house H based on the determination result.

Next, an example of the operation of the fire alarm system 1 described above will be described with reference to FIG. 7. Here, FIG. 7 is a sequence diagram showing an example of the operation of the fire alarm system 1.

First, each of the sensor units 10 installed in the house H transmits the results obtained by sensing to the access point AP (step S11). When the access point AP receives the sensing result from the sensor unit 10, it transfers it to the cloud service device 20 (step S12).

Note that the timing at which the sensor unit 10 (the odor sensor 11, the human sensor 12, and the smoke sensor 13) outputs the sensing results is not particularly limited and can be set arbitrarily. For example, the odor sensor 11, the human sensor 12, and the smoke sensor 13 may be configured to perform sensing at predetermined time intervals and output the sensing results individually or all at once. Further, for example, the odor sensor 11, the human sensor 12, and the smoke sensor 13 may be configured to output sensing results individually or all at once in response to a request from the cloud service device 20.

In the cloud service device 20, when the sensing result acquisition unit 211 acquires the sensing result, the transmission source identification unit 212 identifies the installation location of the transmission source sensor unit 10 (step S13). Furthermore, the situation determining unit 213 determines the occurrence status of a fire around the installation location identified by the source identifying unit 212 based on the acquired sensing results (step S14).

When the situation determination unit 213 determines that there is a suspicion of a fire, the notification control unit 214 sends a notification to the address of the normal notification destination associated with the installation location of the sensor unit 10 (step S15). Further, when the situation determination unit 213 determines that a fire has occurred, the notification control unit 214 sends a notification to the address of the emergency notification destination associated with the installation location of the sensor unit 10 (step S16).

Next, with reference to FIG. 8, the processing executed by the cloud service device 20 will be described. FIG. 8 is a flowchart illustrating an example of a process executed by the cloud service device 20. Note that this process corresponds to steps S13 to S16 in FIG.

First, the sensing result acquisition unit 211 acquires the sensing result transmitted from the sensor unit 10 (step S21). Next, the transmission source identification unit 212 refers to the sensor management file 241 and identifies the installation location corresponding to the sensor identifier included in the sensing result as the installation location of the transmission source (step S22).

Next, the situation determination unit 213 determines the occurrence status of fires around the location of the transmission source based on the sensing results (odor components) of the odor sensor 11 acquired by the sensing result acquisition unit 211 and the situation determination data 243. is determined (step S23). Here, if it is determined that there is no suspicion of fire (step S24; No), the situation determining unit 213 ends this process.

In addition, when it is determined that there is a suspicion of fire (step S24; Yes), the situation determination unit 213 determines whether there is smoke in the vicinity of the installation location of the transmission source based on the sensing result of the smoke detection sensor 13 acquired by the sensing result acquisition unit 211. is detected (step S25). Note that if there is no sensing result of the smoke detection sensor 13 corresponding to the installation location of the transmission source, the situation determination unit 213 sends the sensing result to the corresponding smoke detection sensor 13 via the sensing result acquisition unit 211. It is also possible to request the following.

If smoke is not detected in step S25 (step S25; No), the situation determination unit 213 determines whether the human body is detected near the installation location of the transmission source based on the sensing result of the human sensor 12 acquired by the sensing result acquisition unit 211. It is determined whether the activity level of the movement is less than a threshold value (step S26). Note that if there is no sensing result of the human sensor 12 corresponding to the installation location of the transmission source, the situation determination unit 213 sends the sensing result to the corresponding human sensor 12 via the sensing result acquisition unit 211. It is also possible to request the following.

Here, if the degree of activity of the human body movement is equal to or greater than the threshold (step S26; No), the situation determining unit 213 determines that there is no suspicion of a fire, and ends this process. Further, if the activity level of the human body movement is less than the threshold (step S26; Yes), the report control unit 214 reads the address of the normal report destination associated with the installation location of the transmission source from the report destination management file 242, The notification that a fire sign has been detected is sent to that address (step S27), and the process ends.

On the other hand, if smoke is detected in step S25 (step S25; Yes), the situation determination unit 213 determines that a fire has occurred (step S28). In this case, the report control unit 214 reads the address of the emergency call destination associated with the installation location of the transmission source from the report destination management file 242, reports the occurrence of a fire to that address (step S29), and executes this process. finish.

As described above, according to the present embodiment, the cloud service device 20 uses the sensing results of the sensor unit 10 (the odor sensor 11, the human sensor 12, and the smoke sensor 13) to determine when the sensor unit 10 is installed. The situation of fire occurrence around the installation location is comprehensively determined. The cloud service device 20 then sends a report to a predetermined report destination according to the determination result of the fire occurrence situation. As a result, the cloud service device 20 can comprehensively determine the occurrence of a fire from the sensing results of the odor sensor 11, the human sensor 12, and the smoke sensor 13. It is possible to achieve both this and a reduction in the occurrence of information.

Further, in the cloud service device 20, the notification destination can be switched according to the determination result of the fire occurrence situation, that is, the detection of fire signs or fire occurrence. As a result, the cloud service device 20 can notify the appropriate reporting destination based on the determination result of the fire occurrence situation, and can efficiently prompt the fire to be dealt with.

(Modification 1)
In the above-described embodiment, a learned model that infers the fire probability from the odor components detected by the odor sensor 11 has been described as an example of the situation determination data 243. However, the trained model is not limited to this, and trained models created by other methods may be used.

For example, the sensing results of the odor sensor 11 and the sensing results detected by either or both of the human sensor 12 and the smoke sensor 13 are set as a set of learning data, and the sensing results of this set may be derived from a fire. Data indicating whether or not the data is true as a binary value (1, 0) is prepared as teacher data. Then, by inputting multiple pairs of learning data and teacher data into a creation device that creates a trained model and performing machine learning on the relationship between the learning data and the teacher data, the odor sensor 11 and the human sensor A trained model capable of inferring the fire probability can be created from the sensing results of either or both of the smoke detection sensor 12 and the smoke detection sensor 13.

FIG. 9 is a diagram schematically showing the operation of the situation determination data 243b (trained model) according to this modification. For the trained model created by the above creation method, the sensing results of the odor sensor 11 (odor components), the sensing results of the human sensor 12 (movement activity), and the sensing results of the smoke sensor 13 (smoke (presence or absence), the learned model outputs the probability of fire as an inference result.

In other words, the fire probability output by the situation determination data 243b is a composite determination of the fire occurrence situation in the house H based on the sensing results of the odor sensor 11, the human sensor 12, and the smoke sensor 13. Therefore, the situation determination unit 213 uses the situation determination data 243b to determine the occurrence of a fire, thereby achieving both early detection of fire signs and a reduction in the occurrence of non-fire alarms, as in the above-described embodiment. can be achieved.

Note that the threshold value related to the situation determination can be arbitrarily set without any particular limitation. For example, if the fire probability output by the situation determination data 243b is less than 40%, the situation determination unit 213 determines that there is no suspicion of a fire, and if the fire probability is 40% or more and less than 70%, the situation determination unit 213 determines that a fire is suspected. If it is determined that there is a fire and the fire probability is 70% or more, it may be determined that a fire has occurred.

(Modification 2)
In the embodiment described above, the smoke detection sensor 13 is used as the fire detection sensor, but the present invention is not limited to this, and a heat detection sensor that senses heat may be used as the fire detection sensor. In this case as well, it is possible to comprehensively judge the situation of the fire together with the sensing results of other sensors (odor sensor 11, human sensor 12), so as in the above embodiment, early signs of fire can be detected. It is possible to achieve both detection and a reduction in the occurrence of non-fire alarms.

(Modification 3)
In the embodiment described above, the cloud service device 20 holds the sensor management file 241, the reporting destination management file 242, and the situation determination data 243, but the present invention is not limited to this. For example, the cloud service device 20 can store any or all of the sensor management file 241, the report destination management file 242, and the situation determination data 243 in an external device that the cloud service device 20 can access. A configuration may be adopted in which the above-described processing is performed using the created file.

Although the embodiments of the present invention have been described above, these embodiments are presented as examples and are not intended to limit the scope of the invention. Embodiments can be modified in various ways.

1 Fire alarm system 10 Sensor unit 11 Odor sensor 12 Human sensor 13 Smoke detection sensor 20 Cloud service device 211 Sensing result acquisition unit 212 Source identification unit 213 Situation determination unit 214 Notification control unit

Claims (9)

A fire alarm system comprising a sensor unit and an information processing device connected to the sensor unit via a network,
The sensor unit includes an odor sensor that detects odor, a human sensor that detects human body movement, and a fire detection sensor that detects smoke or heat,
The information processing device includes:
an acquisition unit that acquires sensing results around the installation location where the sensor unit is installed from the sensor unit;
a first determination unit that determines whether there is a suspicion of a fire based on the sensing result of the odor sensor acquired by the acquisition unit;
When the first determination unit determines that there is a suspicion of a fire, the acquisition unit determines the occurrence of a fire around the installation location based on the sensing result of the human sensor or the fire detection sensor acquired by the acquisition unit. a second determination section;
a reporting unit that reports to a preset reporting destination according to the determination result of the second determining unit;
A fire alarm system equipped with The fire alarm system according to claim 1 , wherein the first determination unit determines whether or not there is a suspicion of a fire based on the value of a specific odor component detected by the odor sensor. The first determination unit applies the odor sensor to a trained model created using a plurality of types of odor components serving as learning data and teacher data indicating whether the odor components are derived from a fire. The fire alarm system according to claim 1 , wherein the fire alarm system determines whether or not there is a suspicion of a fire based on an inference result obtained by inputting an odor component included in the sensing result. The second determining unit determines that a fire has occurred when the first determining unit determines that a fire is suspected and when smoke or heat is detected from the sensing result of the fire detection sensor. The fire alarm system described in any one of 3 . The second determination unit determines that even if the first determination unit determines that a fire is suspected, if the amount of movement of the human body represented by the sensing result of the human sensor is greater than or equal to a threshold, The fire alarm system according to any one of claims 1 to 3, which determines that there is no suspicion of fire. The fire alarm system according to any one of claims 1 to 5 , wherein the reporting unit switches the reporting destination according to the determination result of the second determining unit. Obtain sensing results around the installation location where the sensor unit is installed from a sensor unit that includes an odor sensor that detects odor, a motion sensor that detects human body movement, and a fire detection sensor that detects smoke or heat. an acquisition unit to
a first determination unit that determines whether there is a suspicion of a fire based on the sensing result of the odor sensor acquired by the acquisition unit;
When the first determination unit determines that there is a suspicion of a fire, the acquisition unit determines the occurrence of a fire around the installation location based on the sensing result of the human sensor or the fire detection sensor acquired by the acquisition unit. a second determination section;
a reporting unit that reports to a preset reporting destination according to the determination result of the second determining unit;
An information processing device comprising: A fire alarm method executed by an information processing device, the method comprising:
The acquisition unit acquires data from a sensor unit including an odor sensor that detects odor, a motion sensor that detects human body movement, and a fire detection sensor that detects smoke or heat, around the installation location where the sensor unit is installed. Obtain sensing results,
a first determination unit determines whether or not there is a suspicion of a fire based on the sensing result of the odor sensor acquired by the acquisition unit;
When the first determining unit determines that a fire is suspected, the second determining unit determines whether there is a fire in the vicinity of the installation location based on the sensing result of the human sensor or the fire detection sensor acquired by the acquiring unit. Determine the occurrence status of
The reporting unit reports to a preset reporting destination according to the determination result of the second determining unit.
fire alarm methods, including The computer of the information processing equipment,
Obtain sensing results around the installation location where the sensor unit is installed from a sensor unit that includes an odor sensor that detects odor, a motion sensor that detects human body movement, and a fire detection sensor that detects smoke or heat. an acquisition unit to
a first determination unit that determines whether there is a suspicion of a fire based on the sensing result of the odor sensor acquired by the acquisition unit;
When the first determination unit determines that there is a suspicion of a fire, the acquisition unit determines the occurrence of a fire around the installation location based on the sensing result of the human sensor or the fire detection sensor acquired by the acquisition unit. a second determination section;
a reporting unit that reports to a preset reporting destination according to the determination result of the second determining unit;
A program to make it work.

Images