WO2024142592A1 - 異常検知方法、異常検知装置、および、プログラム - Google Patents
異常検知方法、異常検知装置、および、プログラム Download PDFInfo
- Publication number
- WO2024142592A1 WO2024142592A1 PCT/JP2023/039276 JP2023039276W WO2024142592A1 WO 2024142592 A1 WO2024142592 A1 WO 2024142592A1 JP 2023039276 W JP2023039276 W JP 2023039276W WO 2024142592 A1 WO2024142592 A1 WO 2024142592A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- information
- state
- time
- residence
- anomaly detection
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Ceased
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L12/00—Data switching networks
- H04L12/02—Details
- H04L12/12—Arrangements for remote connection or disconnection of substations or of equipment thereof
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
- F24F11/00—Control or safety arrangements
- F24F11/30—Control or safety arrangements for purposes related to the operation of the system, e.g. for safety or monitoring
- F24F11/49—Control or safety arrangements for purposes related to the operation of the system, e.g. for safety or monitoring ensuring correct operation, e.g. by trial operation or configuration checks
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
- F24F11/00—Control or safety arrangements
- F24F11/62—Control or safety arrangements characterised by the type of control or by internal processing, e.g. using fuzzy logic, adaptive control or estimation of values
- F24F11/63—Electronic processing
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B15/00—Systems controlled by a computer
- G05B15/02—Systems controlled by a computer electric
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
- F24F2110/00—Control inputs relating to air properties
- F24F2110/10—Temperature
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
- F24F2130/00—Control inputs relating to environmental factors not covered by group F24F2110/00
- F24F2130/10—Weather information or forecasts
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B2219/00—Program-control systems
- G05B2219/20—Pc systems
- G05B2219/26—Pc applications
- G05B2219/2642—Domotique, domestic, home control, automation, smart house
Definitions
- the present invention provides an anomaly detection method that can detect anomalies based on the actual environment inside and outside the home and the state of the equipment.
- An anomaly detection method executed by an anomaly detection device which acquires a learning model generated by machine learning that discriminates anomalies in an appliance using, as input, for each of a plurality of first times, first presence/absence information indicating the presence or absence of a person in the residence at the first time, first environmental information including at least first time information indicating the first time, and first status information indicating the status of the appliance at the first time, and the learning model receives a notification indicating that the status of the appliance at the residence has changed, and acquires second environmental information including at least second presence/absence information indicating the presence or absence of a person in the residence at a second time when the notification is received and second time information indicating the second time, and inputs second status information indicating the status of the appliance after the change indicated in the notification and the acquired second environmental information into the learning model, thereby executing a detection process to detect an anomaly in the appliance.
- the anomaly detection device can detect anomalies depending on the actual environment inside and outside the residence and the status of the appliance.
- the anomaly detection method which calculates representative environmental information that is representative of the first environmental information, a first difference between the first state information and the first environmental information at the specific time, and a second difference between the representative state information and the representative environmental information, and the fourth state information and the fourth environmental information at the fourth time, (a) if the first difference is smaller than the second difference, generates the control information using the state indicated in the first state information at the specific time as the normal state, and (b) if the first difference is larger than the second difference, generates the control information using the state indicated in the fourth state information as the normal state.
- the anomaly detection device can easily change the appliance in which an anomaly has been detected to a normal state by transmitting control information in both cases where the first environmental information and the fourth environmental information at a specific time match and where they do not match. Specifically, when the first environmental information and the fourth environmental information at a specific time match, the appliance can easily change to a normal state by changing the appliance state to the state at the time of receiving the notification. Furthermore, when the first environmental information and the fourth environmental information at a specific time do not match, the appliance can easily change to a normal state by changing the appliance state to a state relatively close to the state at the time of receiving the notification. In this way, the anomaly detection device can easily change the appliance to a normal state after detecting an anomaly according to the actual environment inside and outside the home and the appliance state.
- the abnormality detection device detects the above change as an abnormality and immediately turns the lighting device off, the user may be surrounded by darkness, causing inconvenience to the user.
- the anomaly detection device can detect anomalies based on the actual environment inside and outside the home and the state of the equipment, and then change the equipment to a normal state while avoiding causing inconvenience to the user.
- the anomaly detection device when the anomaly detection device receives a request from a terminal to change the state of the device, it can detect an anomaly in the terminal by determining the presence or absence of an anomaly in the device using a learning model with environmental information and state information of the device after the change related to the request as input.
- the anomaly detection device can detect an anomaly in the terminal according to the actual environment inside and outside the home and the state of the device.
- the system 1 includes an anomaly detection device 10, a router 20, an air conditioner 22, and a terminal 24.
- the anomaly detection device 10, the router 20, the air conditioner 22, and the terminal 24 are placed in a residence and are communicatively connected via a network 30.
- Network 30 is a network within a residence.
- the communication standard of network 30 may be any type, and may be a wired LAN (Local Area Network) communication standard (such as Ethernet (registered trademark)) or a wireless LAN communication standard (such as Wi-Fi (registered trademark)).
- wired LAN Local Area Network
- Wi-Fi registered trademark
- Air conditioner 22 is an air conditioner, which is an example of equipment (facility equipment or home appliances, etc.) installed in a residence. Note that other equipment installed in the residence, specifically lighting equipment, hot water equipment, electric shutters, electric locks, etc., can also be used as air conditioner 22. Below, supplementary explanations may be given about cases where air conditioner 22 is other equipment.
- the air conditioner 22 when the air conditioner 22 receives a request to change the state of the air conditioner 22 (also called a state change request) via the network 30, the air conditioner 22 changes its state in accordance with the state change request.
- the state change request is transmitted by the terminal 24, for example.
- the air conditioner 22 When the state of the air conditioner 22 changes, it transmits to the anomaly detection device 10 a state notification including information indicating that the state has changed and the state after the change. Furthermore, the air conditioner 22 can repeatedly transmit a state notification including its current state to the anomaly detection device 10 regardless of whether the state has changed. A state notification including its current state does not include information indicating that the state has changed. The air conditioner 22 can repeatedly transmit a state notification including its current state periodically (for example, every hour), and can also transmit a state notification in response to a request transmitted from the anomaly detection device 10. Note that a state notification transmitted by the air conditioner 22 when the state has changed is also referred to as a state change notification. A state change notification is a state notification that includes information indicating that the state of the air conditioner 22 has changed.
- a legitimate request is a request to legitimately change the state of the air conditioner 22.
- a legitimate request is a state change request sent based on a user operating the terminal 24 with the intention of changing the state of the air conditioner 22. For example, a request to change the air conditioner 22 to cooling mode with the intention of lowering the temperature in a room of the residence during the hot summer months while the user is inside the residence is a legitimate request.
- An unauthorized request is a request to unauthorizedly change the state of the air conditioner 22.
- An unauthorized request may be a request sent by unauthorized software (such as a computer virus) installed on the terminal 24.
- An unauthorized request by unauthorized software may be made, for example, with the purpose of causing damage (physical damage or monetary damage, etc.) or discomfort or inconvenience to the user.
- a request to change the air conditioner 22 to heating mode during the hot summer months when no one is present in the residence may be an unauthorized request.
- the state of the air conditioner 22 changed by an unauthorized request may be an abnormal state that is subject to detection by the anomaly detection device 10.
- Fraudulent requests may also be made via the operation reception unit of the air conditioner 22.
- a request to change the operation mode of the air conditioner 22 by a child unintentionally touching the operation reception unit can be considered a fraudulent request because it is contrary to the user's intention.
- the terminal 24 is an information processing terminal connected to the network 30.
- the terminal 24 is, for example, a smartphone, a tablet terminal, or a personal computer.
- the terminal 24 has an OS (Operating System) and software that runs on the OS, and operates based on the operation of the software.
- the terminal 24 can send a state change request that changes the state of the air conditioner 22 to the air conditioner 22 via the network 30.
- the terminal 24 may send a state change request that includes an unauthorized request to the air conditioner 22 based on control by the unauthorized software.
- the configuration of the anomaly detection device 10 will be explained below.
- the communication IF 11 is a communication interface device communicatively connected to the network 30.
- the communication IF 11 receives a communication frame including a status notification from the air conditioner 22, it provides the status notification included in the communication frame to the acquisition unit 12.
- the communication IF 11 includes notification information indicating an abnormality in the air conditioner 22, which is provided by the post-processing unit 15, in a communication frame conforming to the communication standard of the network 30 and transmits the same to the network 30.
- the second environmental information may further include second temperature information indicating the temperature at the residence at the second time, and second weather information indicating the weather at the location of the residence.
- the memory unit 13 is a storage device that stores information.
- the memory unit 13 is a volatile or non-volatile storage device that uses a magnetic disk or semiconductor memory, and specifically, can be realized by a RAM (Random Access Memory), a HDD (Hard Disk Drive), an SSD (Solid State Drive), etc.
- the memory unit 13 stores a state table 131, rule information 132, and a learning model 133.
- the status table 131 is information that includes status information of the air conditioner 22 at multiple times in the past from the present time.
- the status table 131 will be explained in detail later.
- Rule information 132 is rule information that indicates the normal state of the air conditioner 22. Rule information 132 will be explained in detail later.
- the learning model 133 is a learning model that determines an abnormality in the air conditioner 22 using as input, for each of a plurality of past times (corresponding to a first time), presence/absence information (corresponding to first presence/absence information) indicating the presence or absence of people in the residence at the first time, environmental information (corresponding to first environmental information) including at least time information (corresponding to first time information) indicating the first time, and status information (corresponding to first status information) indicating the status of the equipment at the first time, and is a learning model generated by machine learning.
- Examples of the plurality of past times include, but are not limited to, several hundred to several thousand times included in a predetermined period (a period having a length of one year or more).
- the first environmental information may further include, for each of a plurality of first times, first temperature information indicating the temperature at the residence at that first time, and first weather information indicating the weather at the location of the residence at that first time.
- Machine learning is performed, for example, by clustering. That is, in machine learning, by clustering the first environmental information (more specifically, the first presence/absence information and first time information, etc., included in the first environmental information) and the first status information at multiple first times, the relationship between the first environmental information and the first status information is found, and the first environmental information and the first status information when the air conditioner 22 is in an abnormal state (or in a normal state) are grouped.
- An example of a clustering technique is the k-means method.
- the k-means method is a technique for classifying the target information into a given number of clusters using the average of the clusters.
- the generation of the learning model 133 may be performed by the detection unit 14 or by another information processing device.
- the detection unit 14 obtains the learning model 133 by reading it from the detection unit 14.
- the detection unit 14 executes a detection process to detect an abnormality in the air conditioner 22 by inputting into the learning model 133 the status information (corresponding to the second status information) indicating the status of the air conditioner 22 after the change indicated in the status change notification acquired by the acquisition unit 12 and the second environmental information acquired by the acquisition unit 12.
- the recovery process is a process for changing the air conditioner 22 to a normal state (in other words, returning it to normal) when the detection unit 14 detects an abnormality in the air conditioner 22.
- the post-processing unit 15 generates control information for returning the air conditioner 22 to a normal state and transmits the generated control information to the air conditioner 22 via the communication IF 11.
- the post-processing unit 15 when transmitting the control information to the air conditioner 22, the post-processing unit 15 generates a communication frame including the control information and transmits the generated communication frame to the air conditioner 22 via the communication IF 11.
- the control information may be, for example, control information for changing the operating state of the air conditioner 22 to a cooling mode when the air conditioner 22 is operating in a heating mode due to a request to change the air conditioner 22 to the heating mode during the hot summer months. It is assumed that the air conditioner 22 that receives the communication frame changes its operating state to the cooling mode in accordance with the control information.
- the control information may be control information that changes the air conditioner 22 to a normal state after waiting for a standby time.
- the post-processing unit 15 when transmitting the control information to the air conditioner 22, the post-processing unit 15 generates a communication frame including information indicating the standby time and control information, and transmits the generated communication frame to the air conditioner 22 via the communication IF 11. It is assumed that the air conditioner 22 that receives the communication frame waits for the standby time from the time the communication frame is received, and after this standby, changes the operating state to cooling mode in accordance with the control information.
- the standby time may be, for example, about 5 to 10 minutes, but is not limited to this.
- the post-processing unit 15 can decide not to execute the restoration process (in other words, suppress its execution) if, for example, executing the restoration process is likely to cause relatively great damage to the user. For example, if an electric shutter, which is an example of a device, is in a normal state when closed and the electric shutter changes to an open state as an abnormal state, closing the electric shutter through the restoration process may cause the electric shutter to collide with a person or object, causing damage.
- the state of the water heating device is stopped before 23:00:01 on August 10, 2022, and is in the heating state at 23:00:20 on the same day. This corresponds to the case where the changed state indicated in the state change notification received from the water heating device at 23:00:20 on the same day is "heating."
- FIG. 4 a data space having an X1 axis and an X2 axis is shown.
- the X1 axis and the X2 axis correspond to the status information and the environmental information of the air conditioner 22.
- a two-dimensional data space is expressed using two axes in FIG. 4, but in reality, a multidimensional data space can be expressed using more axes.
- the air conditioner 22 operates normally for a relatively long period of time and operates abnormally for a relatively short period of time. This is because, when the air conditioner 22 starts to operate abnormally, the user will notice the abnormal operation and take measures to change the settings of the air conditioner 22, stop the operation of the air conditioner 22, and operate normally (for example, by restoring the settings to the factory settings or by repairing it), and the air conditioner 22 will often resume normal operation. Also, if the abnormality in the operation of the air conditioner 22 is serious, the air conditioner 22 will be unable to continue operating and will stop. Therefore, as described above, the group to which the majority of the multiple points (for example, about 90% or more) belong will be the group that corresponds to the normal state of the air conditioner 22.
- step S104 the detection unit 14 uses the status notification, i.e., the status change notification, received in step S101 to execute processing for detecting an abnormality in the air conditioner 22.
- the detailed processing included in step S104 will be described in detail later.
- step S106 the post-processing unit 15 executes post-processing.
- the detailed processing included in step S106 will be explained in detail later.
- step S111 the detection unit 14 acquires the learning model 133 and also acquires the rule information 132.
- step S113 the detection unit 14 acquires environmental information.
- the environmental information acquired at this time corresponds to the environmental information at the time when the status notification, i.e., the status change notification, was received in step S101.
- step S114 the detection unit 14 executes an abnormality detection process using the rule information 132 (see FIG. 5). For example, if the time to execute step S114 is August and the changed state included in the state change notification received in step S101 is "heating mode," the detection unit 14 determines that rule #1 is not met and detects an abnormality in the air conditioner 22.
- step S115 the detection unit 14 executes anomaly detection using the learning model 133.
- the detection unit 14 detects anomalies in the air conditioner 22 by inputting the changed state information included in the state change notification received in step S101 and the environmental information acquired in step S113 to the learning model 133. More specifically, when a point corresponding to the changed state information and the environmental information is plotted in the data space of FIG. 4, the detection unit 14 determines whether the point is located inside the frame 40 (i.e., belongs to a group corresponding to a normal state) or outside the frame 40 (i.e., does not belong to a group corresponding to a normal state). Then, when the detection unit 14 determines that the point is located outside the frame 40, it detects anomalies in the air conditioner 22. Note that when the point is located on the frame 40, it may be determined that the point is located inside or outside the frame 40. After step S115 is completed, the series of processes shown in FIG. 7 is terminated.
- FIG. 8 is a third flow diagram showing the processing of the anomaly detection device 10 in this embodiment.
- the flow diagram shown in FIG. 8 shows detailed processing included in step S106 in FIG. 6.
- step S121 the post-processing unit 15 determines the content of the processing to be performed as the post-processing. Specifically, the post-processing unit 15 determines whether or not to perform a recovery process to return the air conditioner 22 to a normal state. The post-processing unit 15 also determines whether or not to perform a notification process to notify the user's terminal.
- step S123 the post-processing unit 15 generates control information to be used in the recovery process and transmits the generated control information to the user's terminal.
- the post-processing unit 15 can, for example, generate control information that changes (in other words, returns) the state of the air conditioner 22 to the state before the state change notification was received.
- the post-processing unit 15 obtains state information (corresponding to third state information) that indicates the state of the air conditioner 22 at the time (corresponding to the third time) before the state change notification was received, and generates control information using the state indicated in the third state information as the normal state.
- the terminal 24 changes the state of the air conditioner 22 by sending a state change request to the air conditioner 22 via the server 5.
- FIG. 10 is a block diagram showing the functions of the anomaly detection device 10A in this embodiment.
- the communication IF 11A is a communication interface device communicatively connected to the network N.
- the communication IF 11A receives a communication frame including a status change request from the terminal 24, it provides the status change request included in the communication frame to the acquisition unit 12A.
- the communication IF 11A transmits notification information indicating an abnormality in the air conditioner 22 provided by the post-processing unit 15A to the network N by including it in a communication frame conforming to the communication standard of the network N.
- the abnormality detection device 10A can use the communication IF provided in the server 5 as the communication IF 11A. In this case, it can also be expressed as the abnormality detection device 10A not being provided with the communication IF 11A, but using the communication IF provided in the server 5 as the communication IF 11A.
- the detection unit 14A performs detection processing using at least the learning model 133, and may also perform detection processing using the rule information 132.
- the detection unit 14A can detect an abnormality in the terminal 24 by inputting, into the learning model 133, state information (corresponding to the fifth state information) indicating the state of the air conditioner 22 after the change related to the state change request acquired by the acquisition unit 12A and the environmental information (fifth environmental information) acquired by the acquisition unit 12A. Details of the detection processing using the learning model 133 and the detection processing using the rule information 132 are the same as the processing executed by the detection unit 14 in the first embodiment, and therefore will not be described.
- step S223 the detection unit 14A acquires environmental information.
- the acquired environmental information corresponds to the environmental information at the time when the state change request was received in step S201.
- step S224 the detection unit 14A executes anomaly detection processing using the rule information 132 (see FIG. 5).
- the anomaly detection processing using the rule information 132 is similar to step S113 (see FIG. 7) in the first embodiment.
- step S225 the detection unit 14A executes anomaly detection processing using the learning model 133.
- the anomaly detection processing using the learning model 133 is similar to step S114 (see FIG. 7) in the first embodiment.
- the anomaly detection device 10A of this embodiment can detect an anomaly in the terminal 24 that is attempting to change the state of the equipment, depending on the actual environment inside and outside the home and the state of the equipment.
- each component may be configured with dedicated hardware, or may be realized by executing a software program suitable for each component. Specifically, each component may be realized by being integrated into a home automation controller. Each component may be realized by a program execution unit such as a CPU or processor reading and executing a software program recorded on a recording medium such as a hard disk or semiconductor memory.
- the software that realizes the anomaly detection device of each of the above embodiments is a program such as the following.
Landscapes
- Engineering & Computer Science (AREA)
- Signal Processing (AREA)
- General Engineering & Computer Science (AREA)
- Mechanical Engineering (AREA)
- Chemical & Material Sciences (AREA)
- Combustion & Propulsion (AREA)
- Physics & Mathematics (AREA)
- Fuzzy Systems (AREA)
- Mathematical Physics (AREA)
- Computer Networks & Wireless Communication (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Alarm Systems (AREA)
- Selective Calling Equipment (AREA)
Abstract
Description
本発明者は、「背景技術」の欄において記載した、異常の検知の技術に関し、以下の問題が生じることを見出した。
本実施の形態において、実際の住居内外の環境および機器の状態に応じて異常を検知する異常検知装置などについて説明する。
本実施の形態において、実際の住居内外の環境および機器の状態に応じて異常を検知する異常検知装置などについて、上記実施の形態とは異なる形態を説明する。本実施の形態の異常検知装置は、空気調和機の状態を変化させようとする端末の異常を検知することができる。
5 サーバ
10、10A 異常検知装置
11、11A 通信IF
12、12A 取得部
13 記憶部
14、14A 検知部
15、15A 後処理部
20 ルータ
22 空気調和機
24 端末
30、N ネットワーク
40 枠
41、42 点
131 状態テーブル
132 ルール情報
133 学習モデル
Claims (11)
- 異常検知装置が実行する異常検知方法であって、
複数の第一時刻それぞれについて、当該第一時刻における住居の人の在不在を示す第一在不在情報と、当該第一時刻を示す第一時刻情報とを少なくとも含む第一環境情報と、当該第一時刻における機器の状態を示す第一状態情報とを入力として、前記機器の異常の有無を判別する学習モデルであって、機械学習により生成された学習モデルを取得し、
前記住居における前記機器の状態が変化したことを示す通知を受信し、かつ、前記通知を受信した第二時刻における前記住居における人の在不在を示す第二在不在情報と、前記第二時刻を示す第二時刻情報とを少なくとも含む第二環境情報を取得し、
前記通知に示される変化後の前記機器の状態を示す第二状態情報と、取得した前記第二環境情報とを前記学習モデルに入力することで、前記機器の異常を検知するための検知処理を実行する
異常検知方法。 - 前記検知処理において、前記機器の異常を検知した場合には、
前記機器を正常な状態に変化させる制御情報を生成し、生成した前記制御情報を前記機器に送信する
請求項1に記載の異常検知方法。 - 前記異常検知方法は、さらに、
前記通知を受信する前の第三時刻における前記機器の状態を示す第三状態情報を取得し、
前記制御情報を生成する際には、
前記第三状態情報に示される前記状態を前記正常な状態として用いて、前記制御情報を生成する
請求項2に記載の異常検知方法。 - 前記制御情報を生成する際には、
(1)複数の第一時刻のうち、当該第一時刻における前記第一環境情報と、前記通知を受信した第四時刻における第四環境情報との差異が最も小さい前記第一時刻である特定時刻における前記第一環境情報と、前記第四環境情報とが一致している場合には、前記特定時刻における前記第一状態情報に示される前記状態を前記正常な状態として用いて、前記制御情報を生成し、
(2)前記特定時刻における前記第一環境情報と、前記第四環境情報とが一致していない場合には、
前記複数の第一時刻における前記第一状態情報の代表である代表状態情報、および、前記複数の第一時刻における前記第一環境情報の代表である代表環境情報と、前記特定時刻における前記第一状態情報および前記第一環境情報との第一差異と、前記代表状態情報および前記代表環境情報と、前記第四時刻における第四状態情報および前記第四環境情報との第二差異とを算出し、
(a)前記第一差異が前記第二差異より小さい場合には、前記特定時刻における前記第一状態情報に示される前記状態を前記正常な状態として用いて、前記制御情報を生成し、
(b)前記第一差異が前記第二差異より大きい場合には、前記第四状態情報に示される前記状態を前記正常な状態として用いて、前記制御情報を生成する
請求項2に記載の異常検知方法。 - 前記制御情報は、前記機器を待機時間だけ待機させた後で正常な状態に変化させる制御情報である
請求項2~4のいずれか1項に記載の異常検知方法。 - 前記異常検知方法は、さらに、
前記第二時刻における前記機器の正常な状態を示すルール情報を取得し、
前記第二状態情報が前記ルール情報に適合するか否かを判定し、前記第二状態情報が前記ルール情報に適合しないと判定した場合に、前記機器の異常を検知する
請求項1に記載の異常検知方法。 - 前記第一環境情報は、さらに、複数の第一時刻それぞれについて、当該第一時刻における前記住居における温度を示す第一温度情報と、当該第一時刻における前記住居の位置における天気を示す第一天気情報とを含み、
前記第二環境情報は、さらに、前記第二時刻における、前記住居における温度を示す第二温度情報と、前記住居の位置における天気を示す第二天気情報とを含む
請求項1に記載の異常検知方法。 - 前記機器の異常を検知した場合には、
前記機器の異常を示す情報を、前記住居の居住者が保有する端末に送信する
請求項1に記載の異常検知方法。 - 前記異常検知方法は、さらに、
前記住居における前記機器の状態を変化させる要求を端末から受信した第五時刻における、前記住居における人の在不在を示す第五在不在情報と、前記第五時刻を示す第五時刻情報とを取得し、
前記要求に係る前記機器の変化後の状態を示す第五状態情報と、取得した前記第五在不在情報および前記第五時刻情報とを前記学習モデルに入力することで、前記端末の異常を検知するための検知処理を実行する
請求項1に記載の異常検知方法。 - 複数の第一時刻それぞれについて、当該第一時刻における住居の人の在不在を示す第一在不在情報と、当該第一時刻を示す第一時刻情報とを少なくとも含む第一環境情報と、当該第一時刻における機器の状態を示す第一状態情報とを入力として、前記機器の異常を判別する学習モデルであって、機械学習により生成された学習モデルを取得する検知部と、
前記住居における前記機器の状態が変化したことを示す通知を受信し、かつ、前記通知を受信した第二時刻における前記住居における人の在不在を示す第二在不在情報と、前記第二時刻を示す第二時刻情報とを少なくとも含む第二環境情報を取得する取得部とを備え、
前記検知部は、さらに、
前記通知に示される変化後の前記機器の状態を示す第二状態情報と、取得した前記第二環境情報とを前記学習モデルに入力することで、前記機器の異常を検知するための検知処理を実行する
異常検知装置。 - 請求項1に記載の異常検知方法をコンピュータに実行させるプログラム。
Priority Applications (4)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202380088427.9A CN120419135A (zh) | 2022-12-27 | 2023-10-31 | 异常检测方法、异常检测装置以及程序 |
| JP2024567251A JPWO2024142592A1 (ja) | 2022-12-27 | 2023-10-31 | |
| EP23911376.4A EP4645765A4 (en) | 2022-12-27 | 2023-10-31 | ANOMALY DETECTION METHOD, ANOMALY DETECTION DEVICE, AND PROGRAM |
| US19/246,149 US20250369640A1 (en) | 2022-12-27 | 2025-06-23 | Anomaly detection method, anomaly detection device, and recording medium |
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP2022210028 | 2022-12-27 | ||
| JP2022-210028 | 2022-12-27 |
Related Child Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| US19/246,149 Continuation US20250369640A1 (en) | 2022-12-27 | 2025-06-23 | Anomaly detection method, anomaly detection device, and recording medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2024142592A1 true WO2024142592A1 (ja) | 2024-07-04 |
Family
ID=91717335
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/JP2023/039276 Ceased WO2024142592A1 (ja) | 2022-12-27 | 2023-10-31 | 異常検知方法、異常検知装置、および、プログラム |
Country Status (5)
| Country | Link |
|---|---|
| US (1) | US20250369640A1 (ja) |
| EP (1) | EP4645765A4 (ja) |
| JP (1) | JPWO2024142592A1 (ja) |
| CN (1) | CN120419135A (ja) |
| WO (1) | WO2024142592A1 (ja) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2009141725A (ja) | 2007-12-07 | 2009-06-25 | Daiwa House Industry Co Ltd | 住宅機器遠隔監視サービス提供システム |
| JP2013011987A (ja) * | 2011-06-28 | 2013-01-17 | Toshiba Corp | 異常状態検知装置及び異常状態検知方法 |
| JP2020149407A (ja) * | 2019-03-14 | 2020-09-17 | 株式会社東芝 | 監視システムおよび監視方法 |
| JP2021177319A (ja) * | 2020-05-08 | 2021-11-11 | 東芝ライフスタイル株式会社 | 家電システム |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB2555573B (en) * | 2016-10-21 | 2020-03-25 | Centrica Hive Ltd | HVAC performance monitoring system |
| US20180231603A1 (en) * | 2017-02-15 | 2018-08-16 | Abhay Gupta | Systems and Methods for Detecting Occurence of an Event in a Household Environment |
| US11461441B2 (en) * | 2019-05-02 | 2022-10-04 | EMC IP Holding Company LLC | Machine learning-based anomaly detection for human presence verification |
-
2023
- 2023-10-31 JP JP2024567251A patent/JPWO2024142592A1/ja active Pending
- 2023-10-31 WO PCT/JP2023/039276 patent/WO2024142592A1/ja not_active Ceased
- 2023-10-31 EP EP23911376.4A patent/EP4645765A4/en active Pending
- 2023-10-31 CN CN202380088427.9A patent/CN120419135A/zh active Pending
-
2025
- 2025-06-23 US US19/246,149 patent/US20250369640A1/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2009141725A (ja) | 2007-12-07 | 2009-06-25 | Daiwa House Industry Co Ltd | 住宅機器遠隔監視サービス提供システム |
| JP2013011987A (ja) * | 2011-06-28 | 2013-01-17 | Toshiba Corp | 異常状態検知装置及び異常状態検知方法 |
| JP2020149407A (ja) * | 2019-03-14 | 2020-09-17 | 株式会社東芝 | 監視システムおよび監視方法 |
| JP2021177319A (ja) * | 2020-05-08 | 2021-11-11 | 東芝ライフスタイル株式会社 | 家電システム |
Also Published As
| Publication number | Publication date |
|---|---|
| JPWO2024142592A1 (ja) | 2024-07-04 |
| US20250369640A1 (en) | 2025-12-04 |
| EP4645765A1 (en) | 2025-11-05 |
| CN120419135A (zh) | 2025-08-01 |
| EP4645765A4 (en) | 2026-04-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11722508B2 (en) | Methods, systems, and media for dynamically separating internet of things devices in a network | |
| CN108432282B (zh) | 用于通过无线通信管理电子装置的方法和设备 | |
| US10291714B2 (en) | System and method for using general software to control internet of things (IOT) system | |
| JP5767717B2 (ja) | ネットワーク上のマスター/スレーブ・デバイスの自動構成のためのシステム及び方法 | |
| WO2018153362A1 (zh) | 接入物联网平台的方法、物联网平台和物联网设备 | |
| US20230336576A1 (en) | Threshold value calculation device, anomaly detection device, threshold value calculation method, and anomaly detection method | |
| JP5974931B2 (ja) | 通信装置 | |
| US11962653B2 (en) | Device monitoring method, device monitoring apparatus, and recording medium for transmitting messages containing device control commands based on predetermined conditions | |
| US12342169B2 (en) | Unauthorized communication detection method, unauthorized communication detection device, and recording medium | |
| WO2024142592A1 (ja) | 異常検知方法、異常検知装置、および、プログラム | |
| JP7840952B2 (ja) | マルウェア検知方法、マルウェア検知装置、及びプログラム | |
| CN111726429A (zh) | 一种通信方法、装置、设备及介质 | |
| US20190221096A1 (en) | Security system with occupancy determination based on hvac applications | |
| JP6415925B2 (ja) | 管理装置、管理プログラム、及び管理方法 | |
| WO2016046861A1 (ja) | 制御装置、サーバ装置、制御システム、制御方法および制御プログラム | |
| WO2023282192A1 (ja) | マルウェア検知方法、マルウェア検知装置、及びプログラム | |
| KR101070522B1 (ko) | 스푸핑 공격 탐지 및 차단 시스템 및 방법 | |
| TWI449373B (zh) | 區域網路的管理方法及其裝置 | |
| JP7550975B2 (ja) | 空調機、セキュリティ攻撃対処方法及びプログラム | |
| CN118119035A (zh) | 设备的通信连接方法、电子设备、存储介质和系统 | |
| Wu | A Cybersecurity Framework for Wireless-Controlled Smart Buildings I: Two-Position Control | |
| US20080244063A1 (en) | Automatically detecting managed servers in a network | |
| KR20070074157A (ko) | 홈 네트워크 시스템에서 레지덴셜 게이트웨이의 장애를판단하는 방법 및 시스템 | |
| JP2018063618A (ja) | 不正アクセス監視装置および方法 |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 23911376 Country of ref document: EP Kind code of ref document: A1 |
|
| WWE | Wipo information: entry into national phase |
Ref document number: 2024567251 Country of ref document: JP |
|
| WWE | Wipo information: entry into national phase |
Ref document number: 202380088427.9 Country of ref document: CN |
|
| WWE | Wipo information: entry into national phase |
Ref document number: 2023911376 Country of ref document: EP |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| WWP | Wipo information: published in national office |
Ref document number: 202380088427.9 Country of ref document: CN |
|
| ENP | Entry into the national phase |
Ref document number: 2023911376 Country of ref document: EP Effective date: 20250728 |
|
| ENP | Entry into the national phase |
Ref document number: 2023911376 Country of ref document: EP Effective date: 20250728 |
|
| WWP | Wipo information: published in national office |
Ref document number: 2023911376 Country of ref document: EP |