Estimation system, power meter, distribution board, estimation method, and program

Abstract

To improve the convenience of a user who monitors or manages an electric apparatus.SOLUTION: An inference system 1 includes an acquisition unit 2 and an inference unit 31. The acquisition unit 2 acquires measurement information D1 that includes at least one of a power, a power factor, and a harmonic component level which are computed on the basis of a measurement result of an electricity-related physical amount. The inference unit 31 infers whether or not a prescribed event related to an electric apparatus AP1 which receives a power supply has occurred on the basis of the measurement information D1.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present disclosure generally relates to an estimation system, a power meter, a distribution board, an estimation method, and a program. More particularly, the present disclosure relates to an estimation system that can be applied to monitoring or management of electrical equipment, a power meter including the estimation system, a distribution board including the estimation system, an estimation method, and a program. [Background technology]

[0002] Patent Document 1 discloses a power management system that manages power in multiple user systems, each of which is provided for a specific unit of power consumption. This power management system is composed of smart meters, which are performance data generators installed in each user system, and a management server connected to the smart meters via the Internet, telephone lines, or dedicated lines. Each smart meter measures the amount of power generated or consumed by each consumer during each power usage period in each user system to generate performance data. The management server manages power consumption within the user system based on the performance data.

[0003] This power management system analyzes the temporal fluctuation of total power consumption measured on a consumer-by-consumer basis within the user system, and estimates the individual devices operating within the consumer and their individual power consumption. Based on the estimation results, the power management system also uses an estimation history that records the power status of each device in chronological order to estimate the number of individual devices operating at the start of the prediction from the real-time total power consumption, and further predicts the trend in individual power consumption over the prediction period from the start of the prediction. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] International Publication No. 2020 / 250648 Summary of the Invention [Problem to be solved by the invention]

[0005] Meanwhile, users who monitor or manage the power consumption of individual devices (electrical devices) want to know in real time whether or not an event (such as a malfunction) has occurred in the electrical device. However, manually checking and analyzing information (measurement information) such as power consumption each time can be a time-consuming task.

[0006] The present disclosure has been made in consideration of the above-mentioned reasons, and aims to provide an estimation system, a power meter, a distribution board, an estimation method, and a program that improve convenience for users who monitor or manage electrical equipment. [Means for solving the problem]

[0007] An estimation system according to one aspect of the present disclosure includes an acquisition unit, an estimation unit, a model storage unit; The acquisition unit acquires measurement information including at least one of information on power, power factor, and harmonic component level calculated based on measurement results of physical quantities related to electricity. The estimation unit estimates whether a predetermined event has occurred related to an electrical device to which electricity is supplied, based on the measurement information. The model storage unit stores a trained model that has been machine-learned for the predetermined event. The estimation unit inputs the measurement information to the trained model and estimates whether the predetermined event has occurred based on a result output from the trained model. The result output from the trained model is the probability of the predetermined event occurring.

[0008] A power meter according to one aspect of the present disclosure includes the above-described estimation system, a measurement unit, a calculation unit, and a main unit. The measurement unit measures the physical quantity related to electricity. The calculation unit calculates at least one of the power, the power factor, and the level of the harmonic components based on the measurement result of the measurement unit. The main unit houses the estimation system, the measurement unit, and the calculation unit.

[0009] A distribution board according to one aspect of the present disclosure includes the above-described estimation system, a power meter, and a cabinet. The power meter includes a measurement unit that measures a physical quantity related to the electricity, and a calculation unit that calculates at least one of the power, the power factor, and the level of the harmonic components based on the measurement result of the measurement unit. The cabinet houses the estimation system and the power meter.

[0010] An estimation method according to one aspect of the present disclosure includes: An estimation method executed by one or more processors, the estimation method comprising: The method includes an acquisition processing step and an estimation processing step. In the acquisition processing step, measurement information including at least one of information on power, power factor, and harmonic component level calculated based on measurement results of physical quantities related to electricity is acquired. In the estimation processing step, based on the measurement information, it is estimated whether or not a predetermined event has occurred related to the electrical equipment to which the electricity is supplied. In the estimation processing step, the measurement information is input to a trained model that has been machine-learned for the predetermined event and is stored in a model storage unit, and whether or not the predetermined event has occurred is estimated based on a result output from the trained model. The result output from the trained model is the probability of the predetermined event occurring.

[0011] A program according to one aspect of the present disclosure ,above The above estimation method , to the one or more processors It is a program to execute. [Effects of the Invention]

[0012] The present disclosure has an advantage in that it is possible to improve the convenience of users who monitor or manage electrical devices. [Brief explanation of the drawings]

[0013] [Figure 1] FIG. 1 is a schematic block diagram of an entire system including an estimation system according to an embodiment. [Figure 2] FIG. 2 is a schematic diagram of a notification image including an estimation result of an event in the estimation system. [Figure 3] FIG. 3 is a flowchart illustrating the operation of the estimation system. [Figure 4] FIG. 4 is a schematic block diagram of the entire system including a modified example of the estimation system. DETAILED DESCRIPTION OF THE INVENTION

[0014] (overview) Hereinafter, an estimation system, a power meter, a distribution board, an estimation method, and a program according to an embodiment will be described with reference to the drawings.

[0015] 1 , an estimation system 1 according to one aspect includes an acquisition unit 2 and an estimation unit 31. As an example, in this embodiment, it is assumed that the functions of the estimation system 1 are implemented in a power meter A1 installed in a cabinet 101 of a distribution board 100.

[0016] The acquisition unit 2 acquires measurement information D1 including information on at least one of power, power factor, and harmonic component levels calculated based on measurement results of physical quantities related to electricity. As an example, the "physical quantities related to electricity" are assumed to be current and voltage, particularly AC current and AC voltage in a power supply line that supplies AC power to an electrical device AP1 (load). In this embodiment, a measurement unit A11 (see FIG. 1) of the power meter A1 measures these physical quantities, and a calculation unit A12 (see FIG. 1) of the power meter A1 calculates, for example, power, power factor, and harmonic component levels based on the measurement results. The acquisition unit 2 acquires the measurement information D1 from the calculation unit A12.

[0017] The estimation unit 31 estimates, based on the measurement information D1, whether a predetermined event has occurred regarding the electric appliance AP1 to which electricity is supplied. In this embodiment, the predetermined event is assumed to include at least one of a failure of the electric appliance AP1 and forgetting to turn off the power of the electric appliance AP1 (hereinafter, sometimes simply referred to as forgetting to turn off (forgetting to turn off)). However, the "predetermined event" referred to in this disclosure is not particularly limited as long as it is an event that can be estimated from the measurement information D1. For example, the event may include an event related to an abnormal state of the electric appliance AP1 (such as a failure state, an abnormal state due to aging, an abnormal state due to the intrusion of foreign matter, or a state requiring maintenance due to dirt, etc.). The event may also include an event related to a state that is inconvenient for the user (such as a state in which unnecessary power consumption occurs due to forgetting to turn off the electric appliance). The event may also include an event related to a precursor to the above-mentioned state.

[0018] According to the above-described estimation system 1, it is automatically estimated whether or not a predetermined event has occurred regarding the electric appliance AP1. As a result, the estimation system 1 has an advantage of improving the convenience of a user who monitors or manages the electric appliance AP1.

[0019] Specific examples of a "user" in the present disclosure include a resident or manager of a residential facility (a detached house or an apartment building), or a manager or supervisor of a non-residential facility (a store, office, school, welfare facility, commercial complex, hospital, factory, outdoor facility, etc.). In other words, the estimation system 1 can be applied to the residential facilities and non-residential facilities described above.

[0020] The estimation result by the estimation unit 31 is displayed on a display unit 70 (see FIG. 1) of a dedicated display setting unit 7 connected to the power meter A1 so as to be able to communicate with the power meter A1. The "display unit" in the present disclosure is not limited to the display unit 70 of the dedicated display setting unit 7, but may be a display unit provided in an information terminal such as a laptop computer, a tablet terminal, or a smartphone. For example, the estimation result is also displayed on a display unit 80 (see FIG. 1) of a mobile terminal 8 (smartphone) owned by the user.

[0021] An estimation method according to one aspect includes an acquisition process step and an estimation process step. In the acquisition process step, measurement information D1 is acquired, the measurement information D1 including at least one of information on power, power factor, and harmonic component level calculated based on measurement results of physical quantities related to electricity. In the estimation process step, it is estimated, based on the measurement information D1, whether a predetermined event has occurred related to the electrical appliance AP1 to which the electricity is supplied. The above estimation method has the advantage of improving convenience for a user who monitors or manages the electrical appliance AP1.

[0022] This estimation method is used on a computer system (estimation system 1). That is, this estimation method can also be embodied as a program. A program according to one aspect is a program for causing one or more processors to execute the above estimation method. The program may be recorded on a computer-readable non-transitory recording medium.

[0023] (detail) (1) Overall structure The estimation system 1 according to this embodiment and the entire system (power management system 10) including its peripheral configuration will be described in detail below with reference to FIGS.

[0024] As shown in FIG. 1, the peripheral configuration of the power management system 10 referred to here includes a distribution board 100 equipped with a power meter A1, one or more electrical devices AP1 (four in FIG. 1), a display setting unit 7, a mobile terminal 8, one or more sensors 9 (one in FIG. 1), etc. Note that at least a part of the peripheral configuration may be included in the configuration of the estimation system 1. In the following, as an example, it is assumed that the power management system 10 is applied to a residential facility such as a detached house. In other words, it is assumed that the user is one or more residents living in the residential facility. Therefore, the distribution board 100 and one or more electrical devices AP1, etc. are installed within the residential facility.

[0025] (2) Distribution board As shown in FIG. 1 , the distribution board 100 includes an estimation system 1, a power meter A1, and a cabinet 101. However, in this embodiment, as described above, all of the functions of the estimation system 1 are implemented within the power meter A1. The distribution board 100 also includes a main breaker B0, multiple branch circuits (for example, four branch circuits C1 to C4), and multiple current detectors (for example, four current detectors CT1 to CT4). The cabinet 101 houses therein the main breaker B0, the branch circuits C1 to C4, the current detectors CT1 to CT4, and the power meter A1 in which the functions of the estimation system 1 are implemented. In other words, the cabinet 101 in this embodiment houses the estimation system 1 and the power meter A1.

[0026] The main breaker B0 has a primary terminal, a secondary terminal, and contacts connected to an electrical circuit between the primary terminal and the secondary terminal. The main breaker B0 further has an operating lever for turning the contacts on or off. The main breaker B0 also has a main circuit breaker. The main circuit breaker breaks the contacts when it detects an abnormal condition. An example of an abnormal condition is the occurrence of a ground fault or an overload current.

[0027] A service line W1 (see FIG. 1) connected to the system power supply P1 is electrically connected to the primary terminal of the main breaker B0. This electrically connects the primary terminal of the main breaker B0 to the system power supply P1. In this embodiment, a single-phase three-wire system is assumed as the power distribution system. Therefore, a single-phase three-wire service line W1 connected to the system power supply P1 is electrically connected to the primary terminal of the main breaker B0.

[0028] A trunk line W2 (see FIG. 1) is electrically connected to the secondary terminal of the main breaker B0. In this embodiment, a single-phase three-wire system is assumed as the power distribution system. The trunk line W2 may include a conductive bar for a first voltage pole (L1 phase), a conductive bar for a second voltage pole (L2 phase), and a conductive bar for a neutral pole (N phase).

[0029] Although detailed description is omitted, for example, a current transformer CT0 is provided as a sensor for detecting the current flowing through the main line W2, and the detection result of the current transformer CT0 is input to a watt-hour meter such as a smart meter.

[0030] In this disclosure, a "branch circuit" refers to an individual circuit branched off from the trunk line W2. As shown in FIG. 1, the branch circuits C1 to C4 each include a branch breaker B1 to B4. Each of the branch circuits C1 to C4 has a primary terminal, a secondary terminal, and a contact connected to an electrical path between the primary terminal and the secondary terminal. Each of the branch circuits C1 to C4 further has an operating lever for turning the contact on or off. Each of the branch circuits C1 to C4 also has a breaker. The breaker breaks the contact when it detects an abnormal state. An example of an abnormal state is the occurrence of a leakage current or an overload current in the corresponding branch circuit.

[0031] Primary side terminals of the branch circuits C1 to C4 are electrically connected to the trunk line W2. Secondary side terminals of the branch circuits C1 to C4 can be electrically connected to multiple (four in the illustrated example) electric devices AP1. In the example of Fig. 1, four electric devices AP1 are electrically connected to the branch circuits C1 to C4, respectively. As a result, the system power supply P1 and the four electric devices AP1 are electrically connected via the branch circuits C1 to C4.

[0032] The multiple electrical appliances AP1 are load appliances that consume power, and as shown in FIG. 1 , include a first air conditioning appliance 110, a first lighting appliance 111, a second air conditioning appliance 112, and a second lighting appliance 113. The electrical appliances AP1 shown in FIG. 1 are merely an example, and the type, size, number, etc. of the "electrical appliances" in the present disclosure are not particularly limited. For example, the multiple electrical appliances AP1 may include a power generation device (e.g., a solar power generation device) that generates power and supplies it to other electrical appliances AP1, a power storage device, or a power grid P1. Furthermore, the multiple electrical appliances AP1 may include a power storage device that receives power from the power grid P1 or a power generation device, is charged, and discharges the power to supply it to other electrical appliances AP1.

[0033] As shown in FIG. 1, the multiple current detection units CT1 to CT4 are installed to detect the currents flowing through the multiple branch circuits C1 to C4, respectively. That is, the current detection unit CT1 detects the current input to the first air conditioning device 110. The current detection unit CT2 detects the current input to the first lighting device 111. The current detection unit CT3 detects the current input to the second air conditioning device 112. The current detection unit CT4 detects the current input to the second lighting device 113. The multiple current detection units CT1 to CT4 are assumed to include, for example, current transformers (CTs), but may also include Rogowski coils. The detection results detected by the multiple current detection units CT1 to CT4 are input to the power meter A1 via the multiple CT cables Ca1 to Ca4.

[0034] (3) Power meter The power meter A1 is configured to be able to measure the voltage, current, power, power factor, and integrated power consumption of multiple circuits (e.g., four circuits). The power meter A1 includes an estimation system 1, a measurement unit A11, a calculation unit A12, and a main unit A10. As shown in FIG. 1, the power meter A1 further includes input terminals T1 to T4, a storage unit A13, and a communication unit A14. Note that the estimation system 1, the measurement unit A11, and the calculation unit A12 merely represent functions realized by one or more processors, and do not necessarily represent actual configurations.

[0035] The main unit A10 is a box-shaped housing that houses the estimation system 1, the measurement unit A11, and the calculation unit A12. The main unit A10 is configured to be mountable within a cabinet 101 of a distribution board 100.

[0036] The main unit A10 has, on its front surface, a plurality of (for example, four) sockets into which the connectors of the plurality of CT cables Ca1 to Ca4 can be plugged. When the connectors of the plurality of CT cables Ca1 to Ca4 are plugged into the plurality of sockets, respectively, the terminals of the connectors of the plurality of CT cables Ca1 to Ca4 come into contact with the input terminals T1 to T4. As a result, the plurality of current detection units CT1 to CT4 are electrically connected to the input terminals T1 to T4. The detection results detected by the plurality of current detection units CT1 to CT4 are input to the measurement unit A11 via the input terminals T1 to T4.

[0037] The power meter A1 also has input terminals (for example, screw terminals) to which voltage cables for individually detecting the voltages in the branch circuits C1 to C4 are connected.

[0038] The storage unit A13 is, for example, an electrically rewritable non-volatile semiconductor memory such as a flash memory. The storage unit A13 can store, as "setting information," various settings accepted by user operation input to the display setting unit 7 or the mobile terminal 8. The storage unit A13 can also store history information such as the measurement results of the measurement unit A11, the calculation results of the calculation unit A12, and the estimation results of the estimation system.

[0039] The communication unit A14 has a communication interface for wired communication with the display setting unit 7. The communication unit A14 may also have a communication interface for wireless communication with the display setting unit 7 directly or via a router installed in the residential facility. The wired communication is, for example, wired communication via a twisted pair cable, a dedicated communication line, or a LAN (Local Area Network) cable. The wireless communication is, for example, wireless communication conforming to standards such as Wi-Fi (registered trademark), Bluetooth (registered trademark), ZigBee (registered trademark), or low-power radio that does not require a license (specified low-power radio), or wireless communication such as infrared communication.

[0040] Furthermore, when the mobile terminal 8 is located outside the residential facility, the communication unit A14 has a communication interface for communicating with the mobile terminal 8 via a router in the residential facility and a wide area network such as the Internet. Furthermore, when the mobile terminal 8 is located inside the residential facility, the communication unit A14 has a communication interface for communicating with the mobile terminal 8 via wireless communication compliant with standards such as Wi-Fi (registered trademark).

[0041] Each of the measurement unit A11 and the calculation unit A12 includes a computer system having one or more processors and a memory. At least some of the functions of each of the measurement unit A11 and the calculation unit A12 are realized by the processor of the computer system executing a program recorded in the memory of the computer system. The program may be recorded in the memory, provided via a telecommunications line such as the Internet, or provided by being recorded on a non-transitory recording medium such as a memory card.

[0042] The measurement unit A11 individually acquires detected values ​​(current values) related to the currents flowing through the branch circuits C1 to C4 via the input terminals T1 to T4. The measurement unit A11 also individually acquires detected values ​​(voltage values) related to the voltages in the branch circuits C1 to C4 via the input terminals to which voltage cables are connected. That is, the measurement unit A11 measures the currents and voltages input to the electrical devices AP1 in real time, for example, every minute. In other words, the measurement unit A11 measures physical quantities related to electricity. The measurement results of the measurement unit A11 are input to the calculation unit A12 as needed and stored (memorized) in the memory unit A13.

[0043] The calculation unit A12 calculates a plurality of calculation items from the current and voltage measured by the measurement unit A11 and the setting information stored in the storage unit A13. The calculation items include, for example, instantaneous power, power factor obtained by dividing active power by apparent power, integrated power, and harmonic component levels. The harmonic component levels will be described later.

[0044] The setting information includes information on the current time, CT ratios (current transformation ratios), rated currents of current detection units CT1 to CT4, and phase and wire systems. The setting information also includes name information (e.g., air conditioner 1, air conditioner 2, lighting 1, lighting 2) and type information (e.g., air conditioner or lighting) as individual information about the electric device AP1 connected to each of branch circuits C1 to C4. The setting information also includes manufacturer information, product number (product number or serial number), and installation date as individual information about the electric device AP1. In other words, the setting information includes information for linking each of branch circuits C1 to C4 with the corresponding electric device AP1. The setting information also includes group information specifying groupings for multiple branch circuits C1 to C4 and information specifying the unit time (e.g., by minute, hour, day, or month) for calculating calculation items.

[0045] The calculation unit A12 calculates the above-mentioned multiple calculation items individually for each of the multiple branch circuits C1 to C4. The calculation unit A12 can calculate the above-mentioned multiple calculation items for each minute, hour, day, or month based on the setting information.

[0046] The calculation unit A12 also calculates the above-mentioned multiple calculation items for each group based on the setting information. For example, a user can use the display setting unit 7 or the mobile terminal 8 to set and register the branch circuits C1 and C3, to which the first air conditioning device 110 and the second air conditioning device 112 are connected, as an "air conditioning group." Similarly, a user can use the display setting unit 7 or the mobile terminal 8 to set and register the branch circuits C2 and C4, to which the first lighting device 111 and the second lighting device 113 are connected, as a "lighting group." The storage unit A13 stores group information linking the branch circuits C1 and C3 to the "air conditioning group" and the branch circuits C2 and C4 to the "lighting group." The calculation unit A12 can calculate the above-mentioned multiple calculation items for each "air conditioning group" or "lighting group."

[0047] The calculation unit A12 has a function of calculating the level of harmonic components (e.g., the effective level of harmonic current), which is one of the above-mentioned multiple calculation items, from the signal waveforms of current and voltage measured by the measurement unit A11. The level of harmonic components may increase, similar to the power consumed by the electrical equipment AP1, when the electrical equipment AP1 fails or shows signs of failure. The calculation unit A12 calculates the level of harmonic components individually for each of the multiple branch circuits C1 to C4. The calculation unit A12 can calculate the level of harmonic components every minute, every hour, every day, or every month. The calculation unit A12 can also calculate the level of harmonic components by group.

[0048] In short, the calculation unit A12 calculates the power (instantaneous power and integrated power consumption), power factor, and harmonic component levels based on the measurement results of the measurement unit A11. However, it is not essential for the calculation unit A12 to calculate all of the power, power factor, and harmonic component levels. In other words, the calculation unit A12 calculates at least one of the power, power factor, and harmonic component levels based on the measurement results of the measurement unit A11.

[0049] The calculation unit A12 outputs measurement information D1 (see FIG. 1) including information on the power (instantaneous power and integrated power amount), power factor, and level of harmonic components to the acquisition unit 2 of the estimation system 1.

[0050] (4) Display setting unit and mobile terminal The display setting unit 7 is communicatively connected to the power meter A1 via a connection cable (wired). The display setting unit 7 may be disposed inside the cabinet 101 of the distribution board 100 or outside the cabinet 101. As shown in FIG. 1 , the display setting unit 7 has a display unit 70 and an operation unit 71.

[0051] The display unit 70 includes a display such as a liquid crystal display or an organic electroluminescence (EL) display. The display unit 70 displays (presents) information. The information displayed by the display unit 70 includes, for example, the measurement results of the power meter A1 and the estimation results of the estimation system 1. The display unit 70 can also display a setting screen for inputting the above-mentioned setting information and a screen for notifying various errors. The operation unit 71 includes, for example, one or more push buttons and accepts operation inputs by the user. The display setting unit 7 may further include a speaker, and may output information related to the measurement results of the power meter A1 and the estimation results of the estimation system 1 by voice from the speaker.

[0052] The mobile terminal 8 is connected to the power meter A1 so as to be able to communicate wirelessly. Dedicated application software for communicating with the power meter A1 is installed on the mobile terminal 8.

[0053] As shown in FIG. 1 , the mobile terminal 8 has a display unit 80. The display unit 80 includes a display such as a liquid crystal display or an organic EL display. The display is, for example, a touch panel display that can accept user input. The display unit 80 displays (presents) information. The information displayed on the display unit 80 includes, for example, the measurement results of the power meter A1 and the estimation results of the estimation system 1. The display unit 80 can also display a setting screen for inputting the above-mentioned setting information and a screen for notifying various errors. The mobile terminal 8 may output information related to the measurement results of the power meter A1 and the estimation results of the estimation system 1 by voice from a speaker attached to the mobile terminal 8.

[0054] (5) Sensor One or more sensors 9 (one in FIG. 1 ) detect physical quantities that are the basis of auxiliary information D2 (see FIG. 1 ) including "information other than physical quantities related to electricity" (described later). In this embodiment, the "information other than physical quantities related to electricity" is assumed to be, for example, information that may be related to the operation of the electrical device AP1, but is not particularly limited thereto. Each of the one or more sensors 9 (one in FIG. 1 ) is installed, for example, to monitor the ambient environment of the corresponding electrical device AP1. The one or more sensors 9 are communicably connected to the estimation system 1 via a communication unit A14 of the power meter A1.

[0055] The one or more sensors 9 may include a temperature and humidity sensor that detects temperature and humidity. The temperature and humidity sensor is installed, for example, in the installation space of the electric device AP1 and detects the temperature and humidity around the electric device AP1. Specifically, if the first air conditioner 110 is installed in a space (for example, the living room) in the residential facility, the temperature and humidity sensor is installed in the living room and detects the temperature and humidity around the first air conditioner 110.

[0056] If the electrical device AP1 is an air conditioner, a temperature and humidity sensor provided in a controller that controls the air conditioner may be used as one of the sensors 9. In this case, the communication unit A14 of the power meter A1 can communicate with the controller of the air conditioner.

[0057] The one or more sensors 9 may include a seismic sensor that detects ground vibrations (seismic intensity level of an earthquake). The seismic sensor is installed in the residential facility. The seismic sensor may be installed in the distribution board 100. The one or more sensors 9 may also include a vibration sensor that detects vibrations that may be generated by the electrical apparatus AP1 (for example, vibrations of an internal motor, etc.). The vibration sensor is attached to the electrical apparatus AP1.

[0058] The one or more sensors 9 may include, for example, a gas sensor (which may be an odor sensor) that detects the concentration of a specific gas. The type of specific gas is not particularly limited, but examples include CO, CO2, and combustible gases.

[0059] The one or more sensors 9 may also include an image sensor that captures an image of the installation space of the electric device AP1, or a sound sensor (sound collection microphone) that collects noise and the like in the installation space of the electric device AP1.

[0060] (6) Estimation system The estimation system 1 includes a computer system having one or more processors and a memory. At least some of the functions of the estimation system 1 are realized by the processor of the computer system executing a program recorded in the memory of the computer system. The program may be recorded in the memory, or may be provided via a telecommunications line such as the Internet, or may be provided by being recorded on a non-transitory recording medium such as a memory card.

[0061] 1, the estimation system 1 includes an acquisition unit 2, a processing unit 3, a function unit 4, a model storage unit 5, and a learning unit 6. The estimation system 1 is preferably capable of communicating with each of its peripheral components, such as a display setting unit 7, a mobile terminal 8, and one or more sensors 9. In this embodiment, the functions of the estimation system 1 are implemented in a power meter A1, and therefore the estimation system 1 is capable of communicating with the above-mentioned peripheral components via a communication unit A14 of the power meter A1.

[0062] The acquisition unit 2 acquires measurement information D1 from the calculation unit A12, which includes information on the power, power factor, and harmonic component levels calculated based on the current and voltage measurement results obtained by the measurement unit A11. The measurement information D1 further includes information on the current and voltage measurement results. The acquisition unit 2 outputs the measurement information D1 to the processing unit 3, including information on the electrical device AP1 connected to the branch circuit that is the measurement target among the branch circuits C1 to C4, and the measurement times of the current and voltage. The acquisition unit 2 also stores the measurement information D1 in the storage unit A13 as history information (hereinafter, sometimes referred to as "electrical history"). A portion of the electrical history (for example, history from the most recent to one month ago) may be used to estimate a predetermined event.

[0063] As shown in FIG. 1, the processing unit 3 includes an estimation unit 31, an output processing unit 32, and an extraction unit 33.

[0064] The estimation unit 31 is configured to estimate, based on the measurement information D1, whether or not a predetermined event related to the electrical appliance AP1 to which electricity is supplied has occurred. That is, the estimation unit 31 is configured to execute a process related to the estimation of the predetermined event (hereinafter, may be referred to as an "event estimation process"). The execution timing of the event estimation process is not particularly limited. The event estimation process may be executed automatically at regular time intervals, or may be executed at any timing, for example, when a request to execute the event estimation process is input by the user via the display setting unit 7 or the mobile terminal 8.

[0065] The estimation unit 31 estimates in real time whether a predetermined event has occurred based on the measurement information D1 acquired most recently (for example, at the present time). The estimation unit 31 estimates whether a predetermined event has occurred based on the change over time of the measurement information D1, taking into account not only the most recent measurement information D1 but also the electrical history of the measurement information D1 stored in the memory unit A13.

[0066] In this embodiment, as an example, the estimation unit 31 estimates whether a first event, which is a failure of the electric appliance AP1 (or a sign of a failure), has occurred based on the measurement information D1. The estimation unit 31 also estimates whether a second event, which is leaving the electric appliance AP1 turned off, has occurred based on the measurement information D1. That is, the predetermined event includes a first event and a second event. Hereinafter, when there is no need to distinguish between the first event and the second event, they may be simply referred to as "events."

[0067] The estimation unit 31 estimates that a first event has occurred when at least one of the calculated instantaneous power, integrated power energy, and harmonic component level (effective value level of harmonic current) has increased to exceed a threshold value set corresponding to the instantaneous power, integrated power energy, and harmonic component level, respectively. The estimation unit 31 also estimates that a first event has occurred when the calculated power factor has decreased to below a threshold value set corresponding to the power factor. These threshold values ​​are values ​​that allow determination that the electric device AP1 has failed. These threshold values ​​are preferably set individually according to the type of each electric device AP1.

[0068] These thresholds may be set based on device information input via the display setting unit 7 or the mobile terminal 8. The device information may include, for example, information about the corresponding electrical device AP1 (such as the type of device, power consumption, and rated voltage). Alternatively, the estimation system 1 may automatically set these thresholds based on information such as the instantaneous power, integrated energy, power factor, and effective value of the fundamental wave of the current actually measured by the power meter A1 or the electrical device AP1 in the early stages of installation in the residential facility. Information about these thresholds is stored (memorized) in the memory unit A13.

[0069] The estimation unit 31 also takes into account the electrical history of the measurement information D1 and estimates that a second event has occurred if at least one of the instantaneous power and the accumulated power amount exceeds a predetermined value set corresponding to the instantaneous power and the accumulated power amount over a specific period of time.

[0070] The specific period is assumed to be, for example, a period during which the user is asleep or out and it is highly likely that the corresponding power meter A1 is not in operation. If the facility to which the estimation system 1 is applied is an office, the specific period is assumed to be, for example, a period during which the user is highly likely not working. The specific period may be an hourly period of one day (for example, the time period from midnight to 6:00), a daily period (for example, December 1st to December 10th), or a day of the week (for example, Saturday and Sunday).

[0071] The specific period is set based on input information input through the display setting unit 7 or the mobile terminal 8. The input information includes information directly specifying a time period, a date, a day of the week, or the like. Alternatively, the input information may include information indicating that the user is about to go out, has returned home, is about to go to bed, or has woken up. The estimation system 1 may automatically determine whether the user is out or not based on the location information of the mobile terminal 8, for example, using a global positioning system (GPS). Furthermore, or alternatively, the estimation system 1 may analyze the user's lifestyle patterns (such as bedtimes and time spent outside) and work patterns based on the measurement information D1 measured over a certain period (e.g., one month), and automatically set the specific period based on the analysis results. Information about the specific period is stored (memorized) in the memory unit A13.

[0072] The predetermined value is a value that allows determination that the electric appliance AP1 is in an operating state, not in a power-off state or a standby state. The predetermined value is preferably set individually according to the type of each electric appliance AP1.

[0073] The predetermined value is set based on information input through the display setting unit 7 or the mobile terminal 8. Alternatively, the estimation system 1 may automatically set the predetermined value based on information such as the instantaneous power and the integrated power amount actually measured in the standby state and the operating state of the electric device AP1. Information on the predetermined value is stored (memorized) in the memory unit A13.

[0074] The estimation unit 31 can also estimate whether or not each event has occurred for each electrical device AP1 individually. The estimation unit 31 can also estimate whether or not each event has occurred for each group, such as the above-mentioned "air conditioning group" or "lighting group."

[0075] The estimation unit 31 of this embodiment inputs the measurement information D1 to a trained model M1 (see FIG. 1) and estimates whether a predetermined event has occurred based on the results output from the trained model M1. As shown in FIG. 1, the model storage unit 5 (memory unit) stores the trained model M1 that has been machine-learned for the predetermined event.

[0076] In the present disclosure, a "trained model" refers to a model for which machine learning using training data has been completed. In this embodiment, the trained model M1 is a model for which machine learning using supervised learning has been completed. The trained model M1 is assumed to include, for example, a model using a neural network or a model generated by deep learning using a multilayer neural network. The neural network may include, for example, a convolutional neural network (CNN) or a Bayesian neural network (BNN). The trained model M1 is realized by implementing a trained neural network in an integrated circuit such as an application-specific integrated circuit (ASIC) or a field-programmable gate array (FPGA). The trained model M1 is not limited to a model generated by deep learning. The trained model M1 may also be a model generated by a support vector machine, a decision tree, or the like.

[0077] The trained model M1 is stored in the model storage unit 5. Here, it is assumed that the trained model M1 includes multiple trained models generated for each event, but it may include only one trained model generated to be able to handle multiple events.

[0078] The trained model M1 includes a trained model for a first event (first model M11) and a trained model for a second event (second model M12). Each of the first model M11 and the second model M12 identifies whether or not the corresponding event has occurred, and outputs the identification result from the output layer as, for example, the occurrence probability of the event.

[0079] The first model M11 is trained using training data that is determined to be within the normal range for the instantaneous power, cumulative power consumption, power factor, and harmonic component levels, as well as the amount of change in these values ​​over time, for each type of electrical equipment AP1.

[0080] Here, the estimation unit 31 estimates whether a predetermined event has occurred based on both the measurement information D1 and the auxiliary information D2. Therefore, the first model M11 used by the estimation unit 31 for the event estimation process is trained using data patterns that are determined to be within a normal range, taking into account information other than physical quantities related to electricity, as training data. Factors that may lead to the determination that there is a fault in the electrical device AP1 include, for example, an abnormal increase in the instantaneous power, integrated power consumption, and levels of harmonic components, an abnormal decrease in power factor, an abnormal increase in temperature and humidity, an abnormal increase in vibration, etc.

[0081] The functional unit 4 of this embodiment is configured to acquire auxiliary information D2 including information other than physical quantities related to electricity. The auxiliary information D2 includes at least one of information on temperature, humidity, and vibration. As an example, it is assumed here that the auxiliary information D2 includes all information on the temperature and humidity around the electric appliance AP1 and vibration caused by an internal motor of the electric appliance AP1, etc. Specifically, the functional unit 4 acquires auxiliary information D2 including information on the temperature and humidity around the electric appliance AP1 detected by a temperature and humidity sensor (sensor 9) and information on vibration of the electric appliance AP1 detected by a vibration sensor (sensor 9).

[0082] Correspondence information that associates each sensor 9 with the electrical device AP1 (or any of the branch circuits C1 to C4) whose physical quantity is to be detected by the sensor 9 is stored in the storage unit A13. The correspondence information is set based on information input via the display setting unit 7 or the mobile terminal 8.

[0083] The functional unit 4 outputs auxiliary information D2 including information about the electrical appliance AP1 that was the detection target of the physical quantity and the detection time to the processing unit 3. The functional unit 4 also stores the auxiliary information D2 in the storage unit A13 as a temperature and humidity history (information) and a vibration history (information). A portion of each of the temperature and humidity history and the vibration history (for example, history from the most recent period up to one month ago) may be used to estimate an event.

[0084] The first model M11 receives input data (measurement information D1 including an electrical history, and auxiliary information D2 including a temperature / humidity history and a vibration history) and identifies whether a first event has occurred in the electrical device AP1 that is the measurement target of the measurement information D1. The identification result is output as the occurrence probability of the first event (first occurrence probability). The first occurrence probability may be output without using the auxiliary information D2.

[0085] The estimation unit 31 estimates that there is no failure in the electric appliance AP1 if the first occurrence probability is, for example, less than 50%. The estimation unit 31 estimates that there is a sign of a failure in the electric appliance AP1 if the first occurrence probability is, for example, 50% or more and less than 80%. The estimation unit 31 estimates that there is a failure in the electric appliance AP1 if the first occurrence probability is, for example, 80% or more.

[0086] The second model M12 is trained using training data of data patterns of factors that are judged to be an over-power-off state of the electric appliance AP1 for each type of electric appliance AP1. The factors that are judged to be an over-power-off state include, for example, the electric appliance AP1 being in operation for a long period of time, the electric appliance AP1 being in operation late at night or when the user is out, etc.

[0087] The second model M12 receives input data (measurement information D1 including an electrical history, and auxiliary information D2 including a temperature / humidity history and a vibration history) and identifies whether a second event has occurred in the electrical device AP1 that is the measurement target of the measurement information D1. The identification result is output as a probability that the second event has occurred (second occurrence probability). The second occurrence probability may be output without using the auxiliary information D2.

[0088] For example, assume that the type of the electrical device AP1 to be measured is a lighting device (e.g., the first lighting device 111). If it is estimated from information on the instantaneous power and the integrated energy that the first lighting device 111 is in operation during a late-night time period (a specific period) when it is highly unlikely that lighting devices of this type should be in operation, the second model M12 outputs a high value for the second occurrence probability. Because the specific period may differ depending on the lifestyle patterns of individuals, it is preferable that, after installation in a facility, the estimation system 1 analyzes the lifestyle patterns of individuals from the collected measurement information D1 and auxiliary information D2 and re-learns the second model M12.

[0089] The estimation unit 31 estimates that the electric appliance AP1 has not been forgotten to be turned off if the second occurrence probability is, for example, less than 80%. The estimation unit 31 estimates that the electric appliance AP1 has been forgotten to be turned off if the second occurrence probability is, for example, 80% or more.

[0090] The output processing unit 32 notifies the user of at least a portion of the measurement information D1 and the estimation results of the estimation unit 31. In this embodiment, the output processing unit 32 notifies the user of only the information extracted by the extraction unit 33. The extraction unit 33 is configured to extract predetermined information that satisfies a predetermined condition from the measurement information D1 and the estimation results of the estimation unit 31. In other words, the estimation unit 31 may notify the user of all of the measurement information D1, including the electrical history, and the estimation results of the estimation unit 31 each time it performs the event estimation process. However, this may increase the processing load on the output processing unit 32 and the amount of data communication. By having the extraction unit 33 narrow down the information to be notified to some extent, the processing load on the output processing unit 32 and the amount of data communication can be reduced. Furthermore, for example, information that is likely to be deemed unnecessary by the user is less likely to be notified. In this embodiment, the output processing unit 32 notifies the user of the information extracted by the extraction unit 33 by displaying it on the display unit 70, 80. The output processing unit 32 may notify the user of the information extracted by the extraction unit 33 by outputting it as audio.

[0091] The "predetermined information satisfying the predetermined condition" includes, as an initial setting, information that satisfies the "display request" from the user. The setting information regarding the predetermined condition is stored in the storage unit A13.

[0092] The display setting unit 7 and the mobile terminal 8 can accept an operation input requesting the display of information such as the presence or absence of an event (estimated result), a daily report, a monthly report, or a yearly report. That is, the user may want to know information such as the presence or absence of an event, a daily report, a monthly report, or a yearly report at any time. When the user performs an operation input requesting the display of information such as the presence or absence of an event, a daily report, a monthly report, or a yearly report via the display setting unit 7 or the mobile terminal 8, the extraction unit 33 extracts information (predetermined information) necessary for displaying this information from the measurement information D1 including the electricity history. In some cases, the extraction unit 33 may also extract information from the auxiliary information D2 including the temperature and humidity history and the vibration history in response to a display request from the user.

[0093] Additionally, the "predetermined information satisfying the predetermined condition" initially includes information related to the estimation result of the estimation unit 31 indicating that a predetermined event has occurred. When the estimation result of the estimation unit 31 indicates the occurrence of one or more events, the predetermined information extracted by the extraction unit 33 further includes information on the type of the event that is estimated to have occurred. Furthermore, the predetermined information extracted by the extraction unit 33 further includes information that can identify the electric device AP1 that is the source of the event (for example, a circuit number such as "circuit 1" or name information of the electric device AP1 such as "air conditioning 1" or "lighting 1"), and measurement information D1 of the electric device AP1.

[0094] That is, in the initial setting, when it is estimated that an event will occur, predetermined information is automatically extracted and displayed on the display units 70, 80, even if there is no display request from the user. Note that, since the user may feel annoyed if the estimation results are notified (displayed) frequently on the display units 70, 80, it is preferable to set an upper limit (for example, once) on the number of times the occurrence of the same event is notified in one day.

[0095] The initial settings for the predetermined conditions can be changed based on information input through the display setting unit 7 or the mobile terminal 8. It is preferable that the settings for the predetermined conditions can be changed separately for the display setting unit 7 and the mobile terminal 8.

[0096] The output processing unit 32 generates output information D3 (see FIG. 1 ) based on the information (predetermined information) extracted by the extraction unit 33, and transmits the output information D3 to the display setting unit 7 and the mobile terminal 8 via the communication unit A14. The estimation system 1 is preferably configurable to transmit the output information D3 to either the display setting unit 7 or the mobile terminal 8, or to both, depending on the type of event. For example, the output information D3 regarding the first event (failure) may be transmitted to both the display setting unit 7 and the mobile terminal 8, and the output information D3 regarding the second event (forgetting to turn off) may be transmitted only to the mobile terminal 8.

[0097] In this way, at least a part of the measurement information D1 and the estimation result of the estimation unit 31 is transmitted to the display setting unit 7 and the mobile terminal 8 and displayed on the display units 70 and 80. Therefore, the user can visually know at least a part of the measurement information D1 and the estimation result of the estimation unit 31. As a result, user convenience is improved.

[0098] The output processing unit 32 preferably displays on the display units 70, 80 the measurement information D1 and the estimation results of the estimation unit 31 in a manner that allows only specific information to be viewed. Setting information regarding the specific information is stored in the storage unit A13. In the initial setting, the specific information includes, for example, the type of event that is estimated to have occurred and the name information of the electrical device AP1 that corresponds to the source of the event. The setting regarding the specific information can be changed based on information input through the display setting unit 7 or the mobile terminal 8. It is preferable that the setting regarding the specific information can be changed separately on the display setting unit 7 and the mobile terminal 8.

[0099] The output processing unit 32 generates and outputs output information D3 including information on a notification image S1 (see FIG. 2) that has been processed, for example, by graphing measurement information D1 including an electrical history to make it easier for the user to understand changes over time. As a result, the notification image S1 is displayed on the display units 70, 80 of the display setting unit 7 and the mobile terminal 8. The notification image S1 may also include auxiliary information D2 as necessary.

[0100] 2 shows an example of the notification image S1. The notification image S1 includes a graph image S2 showing a bar graph of the (accumulated) amount of power consumed for each hour, with the horizontal axis representing time from midnight to 11:00 PM. The notification image S1 also includes an announcement image S3 (specific information) that is displayed as a pop-up on the graph image S2. The announcement image S3 includes an image area showing the type of event that is estimated to have occurred (forgotten to turn off (forgotten to turn off in FIG. 2)), the location information (1F living room) and name information (lighting 1) of the electrical device AP1 that corresponds to the source of the event, and the estimated time (7:00 PM).

[0101] The notification image S1 in FIG. 2 is an example of a notification of an inference result that a second event (forgetting to turn off) has occurred in the first lighting device 111 connected to the branch circuit C2 and installed in the living room on the first floor. In the example of the notification image S1, the inference system 1 receives information that the user has been out since 6:00 PM, for example, but determines that the first lighting device 111 is operating based on the accumulated power consumption at 7:00 PM, and notifies the user of the inference result that a second event has occurred. For example, the user can learn that "Light 1" in the living room on the first floor has been forgotten to be turned off by viewing an announcement image S3 displayed as a pop-up on the display unit 80 of the portable terminal 8 that the user carries. The portable terminal 8 may notify the user of the inference result by email or push notification.

[0102] It is preferable that the output processing unit 32 displays, on the display units 70 and 80, information (such as the most recent accumulated power consumption) that is the cause of the estimation unit 31 estimating that an event has occurred from the measurement information D1 by changing the display color or the like so that the information is visually easy to understand. In the example of Fig. 2, the bar indicating the accumulated power consumption at 7 pm is displayed in a different color (dot hatching in Fig. 2) from the bars indicating the accumulated power consumption at other times.

[0103] In this way, by displaying only specific information closely related to the event (announcement image S3) in a viewable manner (pop-up display manner) on the display units 70, 80, important information is more easily notified to the user. Therefore, user convenience is further improved compared to, for example, a case where information that is likely to be deemed unnecessary by the user is displayed.

[0104] The extraction unit 33 may use a trained model (a model different from the trained model M1) that has been machine-learned about the user's preferences to determine which information to extract as the predetermined information. The output processing unit 32 may use the trained model (a model different from the trained model M1) to determine which information to display on the display units 70 and 80 in a viewable manner.

[0105] The learning unit 6 is configured to re-learn the trained model M1 based on the measurement information D1 acquired by the acquisition unit 2. The learning unit 6 analyzes, for example, a "specific period," such as a time of day or a day of the week, during which the user is likely not using the electrical appliance AP1, based on the measurement information D1 including the electricity history and operation input to the display setting unit 7 or the mobile terminal 8. For example, if a new "specific period" is found other than a late-night period, the learning unit 6 re-learns the second model M12 so as to output a high value of the second occurrence probability when the electrical appliance AP1 is operating during the "specific period." Conversely, if the learning unit 6 determines from the above analysis that the user is likely to normally use the electrical appliance AP1 during late-night hours (e.g., using the air conditioner in sleep mode), the learning unit 6 re-learns the second model M12 so as to output a low value of the second occurrence probability even when the electrical appliance AP1 is operating during late-night hours. Re-learning the trained model M1 improves the reliability of the estimation result by the estimation unit 31.

[0106] (7) Event estimation A series of flows related to event estimation executed by the estimation system 1 will be described below with reference to Fig. 3. Note that the flowchart shown in Fig. 3 is merely an example of the flow of event estimation according to the present disclosure, and the order of processes may be changed as appropriate, and processes may be added or omitted as appropriate.

[0107] The power meter A1 acquires, in the measuring unit A11, detected values ​​relating to the currents flowing through the branch circuits C1 to C4 and detected values ​​relating to the voltages in the branch circuits C1 to C4 (step ST1: measuring currents and voltages).

[0108] The power meter A1 calculates a plurality of calculation items (instantaneous power, power factor, integrated power amount, effective value of harmonic current, etc.) based on the measured current and voltage in the calculation unit A12 (step ST2: calculation of power, etc.).

[0109] The estimation system 1 acquires, in the acquisition unit 2, measurement information D1 including the measurement results (current and voltage) of the measurement unit A11 and information such as instantaneous power, power factor, integrated power amount, and effective value of harmonic current calculated based on the measurement results (step ST3). In other words, the estimation method in the present disclosure includes an acquisition processing step of acquiring measurement information D1 including information on the power, power factor, and harmonic component levels calculated based on the measurement results of physical quantities related to electricity.

[0110] The estimation system 1 acquires auxiliary information D2 in the functional unit 4, which includes information on the temperature and humidity around the electric appliance AP1 detected by the temperature and humidity sensor, and information on the vibration of the electric appliance AP1 detected by the vibration sensor (step ST4).

[0111] The estimation system 1 uses the trained model M1 to acquire the occurrence probability of the first event (first occurrence probability) and the occurrence probability of the second event (second occurrence probability) from the measurement information D1, auxiliary information D2, etc., in the estimation unit 31 (step ST5). Then, the estimation system 1 uses the trained model M1 to estimate whether or not each event will occur from the first occurrence probability and the second occurrence probability (step ST6). In other words, the estimation method in the present disclosure includes an estimation processing step of estimating whether or not a predetermined event related to the electric appliance AP1 to which electricity is supplied has occurred, based on the measurement information D1.

[0112] When the estimation unit 31 of the estimation system 1 estimates that at least one of the first event and the second event has occurred (ST6: Yes), the output processing unit 32 generates output information D3 including the estimation result (step ST7). The estimation system 1 transmits the output information D3 to the display setting unit 7 and the mobile terminal 8 via the communication unit A14, and causes the display units 70 and 80 to display that an event has occurred (step ST8: notification of event occurrence). Note that when the estimation unit 31 estimates that neither the first event nor the second event has occurred (ST6: No), the estimation system 1 does not notify the user and ends the series of processes related to event estimation.

[0113] In this way, the estimation system 1 automatically estimates whether or not each event related to the electric appliance AP1 has occurred. This increases the possibility of satisfying a desire to know in real time whether or not an event (such as a malfunction or forgetting to turn off) related to the electric appliance AP1 has occurred. As a result, the estimation system 1 has the advantage of improving the convenience of a user who monitors or manages the electric appliance AP1.

[0114] In addition, the estimation system 1 acquires auxiliary information D2 including information other than physical quantities related to electricity, and estimates whether each event has occurred based on both the measurement information D1 and the auxiliary information D2, thereby improving the reliability of the estimation results.

[0115] Furthermore, in the estimation system 1, the auxiliary information D2 includes at least one of information on temperature, humidity, and vibration, so that the auxiliary information D2 can be acquired by a simple method such as using a sensor, and the reliability of the estimation results is further improved.

[0116] In particular, the estimation system 1 uses the trained model M1 to estimate whether each event has occurred, further improving the reliability of the estimation results. In addition, the use of the trained model M1 tends to reduce the time spent on the event estimation process, increasing the possibility of notifying the user of the event estimation results in more real time.

[0117] (8) Variations The above embodiment is merely one of various embodiments of the present disclosure. The above embodiment can be modified in various ways depending on the design, etc., as long as the object of the present disclosure can be achieved. Furthermore, functions similar to those of the estimation system 1 according to the above embodiment may be embodied in an estimation method, a computer program, a non-transitory recording medium on which a computer program is recorded, or the like.

[0118] Modifications of the above embodiment are listed below. The modifications described below can be applied in appropriate combinations.

[0119] The estimation system 1 in the present disclosure includes a computer system. The computer system is primarily composed of a processor and memory as hardware. The processor executes a program stored in the memory of the computer system to realize the functions of the estimation system 1 in the present disclosure. The program may be pre-stored in the memory of the computer system, provided via a telecommunications line, or provided in a non-transitory recording medium such as a memory card, optical disk, or hard disk drive that is readable by the computer system. The processor of the computer system is composed of one or more electronic circuits, including a semiconductor integrated circuit (IC) or a large-scale integrated circuit (LSI). The integrated circuits, such as ICs and LSIs, are referred to by different names depending on the degree of integration, and include integrated circuits called system LSIs, very large-scale integrations (VLSIs), and ultra-large-scale integrations (ULSIs). Furthermore, field-programmable gate arrays (FPGAs), which are programmed after the LSI is manufactured, or logic devices that allow the reconfiguration of internal connections or circuit partitions within the LSI can also be used as processors. Multiple electronic circuits may be integrated on a single chip or distributed across multiple chips. The multiple chips may be integrated into one device or distributed across multiple devices. The computer system referred to here includes a microcontroller having one or more processors and one or more memories. Therefore, the microcontroller also comprises one or more electronic circuits, including semiconductor integrated circuits or large-scale integrated circuits.

[0120] Furthermore, it is not essential that the multiple functions of the estimation system 1 are concentrated in one housing. For example, the components of the estimation system 1 may be distributed across multiple housings.

[0121] Conversely, as in the above embodiment, multiple functions of the estimation system 1 may be integrated into one housing. Furthermore, at least some of the functions of the estimation system 1, for example, some of the functions of the estimation system 1 may be realized by cloud computing or the like.

[0122] Specifically, at least some of the functions of the estimation unit 31 may be realized by cloud computing. For example, FIG. 4 shows a modified example of the estimation system 1 in the present disclosure. In the modified example shown in FIG. 4, all of the functions of the estimation unit 31 and the learning unit 6 are provided in a cloud server SV1 that constitutes part of the cloud computing. The trained model M1 is also provided in the cloud server SV1. Other functions of the estimation system 1 are provided in the power meter A1, as in the above-described embodiment.

[0123] The cloud server SV1 may be composed of one or more server devices. The cloud server SV1 is installed, for example, outside the facility where the power meter A1 is installed. The cloud server SV1 can communicate with the power meter A1 within the facility via a wide area network such as the Internet. The power meter A1 transmits information necessary for event estimation to the cloud server SV1 and receives the estimation result of the estimation unit 31 in the cloud server SV1 from the cloud server SV1.

[0124] 4, a configuration is adopted in which the functions of the estimation unit 31 are provided in a cloud server SV1 and are realized by cloud computing. Therefore, compared to a case in which all of the functions of the estimation unit 31 are implemented in a user terminal (power meter A1), for example, the processing load on the user terminal can be reduced.

[0125] The cloud server SV1 may be configured to be able to communicate with multiple terminals (power meters A1) in multiple facilities, and the estimation unit 31 may perform event estimation processing individually within each facility. The learning unit 6 of the cloud server SV1 may aggregate and analyze measurement information D1, auxiliary information D2, etc. from various users' facilities and re-learn the trained model M1.

[0126] In the above-described embodiment, the estimation unit 31 of the estimation system 1 estimates the occurrence or non-occurrence of each event using the trained model M1. However, the use of the trained model M1 is not essential for the estimation system 1. The estimation unit 31 may estimate the occurrence or non-occurrence of each event by substituting input data (measurement information D1, auxiliary information D2, etc.) into a predetermined arithmetic formula to find the occurrence probability of each event.

[0127] In the above-described embodiment, the predetermined event has been described as including a first event (failure) and a second event (forgetting to turn off). However, the predetermined event is not limited to these, and the number of predetermined events is not limited to two, but may include one, or three or more events. For example, if the electric device AP1 is a power generation device or a power storage device, the estimation system 1 may estimate, from the measurement information D1, whether an "event" has occurred, such as the arrival of a maintenance period or a part replacement period for the electric device AP1.

[0128] Incidentally, if the facility is a store or an office, an access control system, a fire alarm system, or the like may be installed. The functional unit 4 of the estimation system 1 may acquire information other than the "electricity-related physical quantity" from these other systems.

[0129] For example, when the functional unit 4 acquires information regarding the presence or absence of people in a specified management area such as a conference room from the entry / exit management system as auxiliary information D2, the reliability of the estimation unit 31's estimation of the second event (electrical device AP1 in the management area being left on) is improved.

[0130] Furthermore, for example, when the function unit 4 acquires information on the presence or absence of a fire within a predetermined management area from a fire alarm system as auxiliary information D2, this information can be used as information for determining whether or not a temperature rise in the temperature and humidity sensor (sensor 9) is caused by a fire. As a result, the reliability of the estimation of the first event (failure) and the second event (forgetting to turn off) by the estimation unit 31 is improved.

[0131] (summary) The above-described embodiments and the like disclose the following aspects.

[0132] An estimation system (1) according to a first aspect includes an acquisition unit (2) and an estimation unit (31). The acquisition unit (2) acquires measurement information (D1) including at least one of information on power, power factor, and harmonic component level calculated based on measurement results of physical quantities related to electricity. The estimation unit (31) estimates, based on the measurement information (D1), whether a predetermined event has occurred related to an electrical appliance (AP1) to which electricity is supplied.

[0133] According to the above aspect, whether or not a predetermined event related to the electric device (AP1) has occurred is automatically estimated by the estimation system (1). As a result, the estimation system (1) has an advantage of improving the convenience of a user who monitors or manages the electric device (AP1).

[0134] Regarding the estimation system (1) according to the second aspect, in the first aspect, the predetermined event includes at least one of a failure of the electric device (AP1) and forgetting to turn off the electric device (AP1).

[0135] According to the above aspect, the occurrence of a failure of the electric device (AP1) or forgetting to turn off the electric device (AP1) is automatically estimated by the estimation system (1), thereby improving user convenience.

[0136] The estimation system (1) according to a third aspect is the first or second aspect, further comprising a functional unit (4) that acquires auxiliary information (D2) including information other than physical quantities related to electricity. The estimation unit (31) estimates whether a predetermined event has occurred based on both the measurement information (D1) and the auxiliary information (D2).

[0137] According to the above aspect, the reliability of the estimation result by the estimation unit (31) is improved.

[0138] Regarding the estimation system (1) according to the fourth aspect, in the third aspect, the auxiliary information (D2) includes at least one of information on temperature, humidity, and vibration.

[0139] According to the above aspect, the auxiliary information (D2) can be acquired by a simple method such as using a sensor, and the reliability of the estimation result by the estimation unit (31) is further improved.

[0140] The estimation system (1) according to a fifth aspect is any one of the first to fourth aspects, and further includes an output processing unit (32) that notifies at least a part of the measurement information (D1) and the estimation result of the estimation unit (31).

[0141] According to the above aspect, the user can know at least a part of the measurement information (D1) and the estimation result of the estimation unit (31), which further improves user convenience.

[0142] Regarding the estimation system (1) according to the sixth aspect, in the fifth aspect, the output processing unit causes the display unit (70, 80) to display only specific information from the measurement information (D1) and the estimation results of the estimation unit (31) in a viewable manner.

[0143] According to the above aspect, user convenience is further improved compared to, for example, a case where information that is likely to be determined as unnecessary for the user is displayed.

[0144] The estimation system (1) according to a seventh aspect is the fifth or sixth aspect, and further includes an extraction unit (33) that extracts predetermined information that satisfies predetermined conditions from the measurement information (D1) and the estimation result of the estimation unit (31). The output processing unit (32) notifies only the predetermined information extracted by the extraction unit (33).

[0145] According to the above aspect, the processing load on the output processing unit 32 is reduced. Also, for example, information that is likely to be deemed unnecessary by the user is less likely to be notified.

[0146] Regarding the estimation system (1) according to the eighth aspect, in the seventh aspect, the predetermined information that satisfies the predetermined condition includes information about the estimation result of the estimation unit (31) that indicates that a predetermined event has occurred.

[0147] According to the above aspect, the processing load on the output processing unit 32 is further reduced. Also, for example, information that is likely to be deemed unnecessary by the user is less likely to be notified.

[0148] With regard to the estimation system (1) according to the ninth aspect, in any one of the first to eighth aspects, at least a part of the functions of the estimation unit (31) is realized by cloud computing.

[0149] According to the above aspect, the processing load on the user's terminal or the like can be reduced compared to when all of the functions of the estimation unit (31) are implemented in the user's terminal or the like.

[0150] The estimation system (1) according to a tenth aspect is any one of the first to ninth aspects, further comprising a model storage unit (5) that stores a trained model (M1) that has been machine-learned for a predetermined event. The estimation unit (31) inputs measurement information (D1) to the trained model (M1) and estimates whether the predetermined event has occurred based on a result output from the trained model (M1).

[0151] According to the above aspect, the reliability of the estimation result by the estimation unit (31) is improved.

[0152] The estimation system (1) according to the eleventh aspect is the tenth aspect, further comprising a learning unit (6) that re-learns the trained model (M1) based on the measurement information (D1) acquired by the acquisition unit (2).

[0153] According to the above aspect, the trained model (M1) is retrained, thereby further improving the reliability of the estimation result by the estimation unit (31).

[0154] A power meter (A1) according to a twelfth aspect includes the estimation system (1) according to any one of the first to eleventh aspects, a measurement unit (A11), a calculation unit (A12), and a main unit (A10). The measurement unit (A11) measures a physical quantity related to electricity. The calculation unit (A12) calculates at least one of power, a power factor, and a level of a harmonic component based on a measurement result of the measurement unit (A11). The main unit (A10) accommodates the estimation system (1), the measurement unit (A11), and the calculation unit (A12).

[0155] According to the above aspect, it is possible to provide a power meter (A1) that can improve convenience for a user who monitors or manages an electric device (AP1).

[0156] A distribution board (100) according to a thirteenth aspect includes the estimation system (1) according to any one of the first to eleventh aspects, a power meter (A1), and a cabinet (101). The power meter (A1) includes a measurement unit (A11) that measures a physical quantity related to electricity, and a calculation unit (A12) that calculates at least one of power, a power factor, and a level of a harmonic component based on a measurement result of the measurement unit (A11). The cabinet (101) accommodates the estimation system (1) and the power meter (A1).

[0157] According to the above aspect, it is possible to provide a distribution board (100) that can improve convenience for a user who monitors or manages an electric device (AP1).

[0158] An estimation method according to a fourteenth aspect includes an acquisition process step and an estimation process step. In the acquisition process step, measurement information (D1) including at least one of information on power, power factor, and harmonic component level calculated based on measurement results of physical quantities related to electricity is acquired. In the estimation process step, based on the measurement information (D1), it is estimated whether or not a predetermined event has occurred related to an electrical device (AP1) to which electricity is supplied.

[0159] According to the above aspect, it is possible to provide an estimation method that can improve convenience for a user who monitors or manages the electric device (AP1).

[0160] A program according to a fifteenth aspect is a program for causing one or more processors to execute the estimation method according to the fourteenth aspect.

[0161] According to the above aspect, it is possible to provide a function that improves convenience for a user who monitors or manages the electric device (AP1).

[0162] The configurations according to the second to eleventh aspects are not essential for the estimation system (1) and can be omitted as appropriate. [Explanation of symbols]

[0163] 1. Estimation System 2 Acquisition part 31 Estimation part 32 Output Processing Section 33 Extraction part 4 Functional Section 5 Model storage section 6. Learning Department 70,80 Display section 100 Distribution board 101 Cabinet A1 Power Meter A10 main unit A11 Measurement section A12 Arithmetic section AP1 Electrical Equipment D1 Measurement Information D2 Auxiliary information M1 pre-trained model

Claims

1. an acquisition unit that acquires measurement information including at least one of information on power, power factor, and harmonic component level calculated based on measurement results of physical quantities related to electricity; an estimation unit that estimates whether a predetermined event related to the electrical appliance to which electricity is supplied has occurred based on the measurement information; a model storage unit for storing a trained model obtained by machine learning for the predetermined event; Equipped with the estimation unit inputs the measurement information to the trained model and estimates whether the predetermined event has occurred based on a result output from the trained model; The result output from the trained model is the occurrence probability of the predetermined event. Estimation system.

2. The predetermined event includes at least one of a failure of the electrical device and forgetting to turn off the power of the electrical device. The estimation system of claim 1 .

3. Further, a functional unit for acquiring auxiliary information including information other than the physical quantity related to the electricity is provided, the estimation unit estimates whether the predetermined event has occurred based on both the measurement information and the auxiliary information. The estimation system according to claim 1 or 2.

4. The auxiliary information includes at least one of temperature, humidity, and vibration information. The estimation system according to claim 3 .

5. an output processing unit that notifies at least a part of the measurement information and the estimation result of the estimation unit; The estimation system according to any one of claims 1 to 4.

6. the output processing unit notifies the user by displaying only specific information out of the measurement information and the estimation result of the estimation unit on a display unit in a viewable manner. The estimation system according to claim 5 .

7. an extraction unit that extracts predetermined information that satisfies a predetermined condition from the measurement information and the estimation result of the estimation unit, the output processing unit notifies only the predetermined information extracted by the extraction unit. The estimation system according to claim 5 or 6.

8. the predetermined information that satisfies the predetermined condition includes information on an estimation result of the estimation unit that indicates that the predetermined event has occurred. The estimation system according to claim 7 .

9. At least a part of the functions of the estimation unit is realized by cloud computing. The estimation system according to any one of claims 1 to 8.

10. The method further comprises a learning unit that re-learns the trained model based on the measurement information acquired by the acquisition unit. The estimation system according to any one of claims 1 to 9.

11. An estimation system according to any one of claims 1 to 10; a measurement unit for measuring the electrical physical quantity; a calculation unit that calculates at least one of the power, the power factor, and the level of the harmonic components based on the measurement result of the measurement unit; a main body unit that houses the estimation system, the measurement unit, and the calculation unit; Equipped with Power meter.

12. An estimation system according to any one of claims 1 to 10; a power meter including a measuring unit that measures the physical quantity related to electricity, and a calculating unit that calculates at least one of the power, the power factor, and the level of the harmonic components based on the measurement result of the measuring unit; a cabinet that houses the estimation system and the power meter; Equipped with Distribution board.

13. An estimation method executed by one or more processors, comprising: an acquisition processing step of acquiring measurement information including at least one of information on power, power factor, and harmonic component level calculated based on measurement results of physical quantities related to electricity; an estimation processing step of estimating whether or not a predetermined event related to the electrical appliance to which electricity is supplied has occurred based on the measurement information; Including, In the estimation processing step, the measurement information is input to a trained model that has been machine-learned for the predetermined event and is stored in a model storage unit, and based on a result output from the trained model, it is estimated whether or not the predetermined event has occurred; The result output from the trained model is the occurrence probability of the predetermined event. Estimation method.

14. A program for causing one or more processors to execute the estimation method described in claim 13.

Images