JP6470497B2 - Action determination system, security system and resident monitoring system - Google Patents

Description

  The present invention relates to a behavior determination system for determining a resident's behavior. The present invention also relates to a security system and a resident watching system using the determination result of the behavior determination system.

  In order to use it for security systems that monitor homes to ensure the safety of residents, and for resident monitoring systems that detect changes in residents who are highly protected due to age or disability, etc. A system for determining the presence / absence of a resident has been developed.

  For example, in Patent Literature 1, based on power data indicating the amount of power consumption in a home output from a power sensor, and door opening / closing timing data indicating an open / closed state of a home door output from a door open / close detection sensor. A system for determining the presence / absence of a resident in the home has been proposed.

  In the system proposed in Patent Document 1, the presence / absence determination of a resident is improved by using door opening / closing timing data in addition to power data. Specifically, in this system, when the power consumption is small but the door is not opened or closed, the resident presence / absence determination accuracy is improved by determining that the resident is at home.

JP 2013-192370 A

  The system proposed in Patent Document 1 determines the presence / absence of a resident and outputs data indicating the determination result (existence determination result). However, only the data indicating the presence / absence of the resident may not be enough to grasp the resident's situation, which may cause a problem.

  Specifically, for example, in the security system, even if it is detected that the resident is at home based on data indicating the presence / absence of the resident, the resident is in an unprotected state (for example, sleeping) It is impossible to detect whether or not actions for ensuring safety, such as evacuation and notification, cannot be performed promptly, such as during bathing. For this reason, in this security system, it is assumed that the resident who is at home is always in an unprotected state. Or restricting actions. Conversely, this security system can endanger the occupant by not performing sufficient monitoring, assuming that the resident at home is not always vulnerable.

  Further, for example, even if data indicating the presence / absence of a resident is obtained in the resident monitoring system, it is difficult to immediately detect the change of the resident. Note that this resident monitoring system may detect that the resident has changed due to the fact that the resident's home or absence is unnaturally long. Then, a serious delay may occur from the time when the resident changes to the time when the change is detected.

  Then, an object of this invention is to provide the action determination system which can output data with high utility value.

  In order to achieve the above object, an action determination system according to the present invention includes a database that records energy consumption data indicating temporal fluctuations in energy consumption, which is the amount of energy consumed at home, and the energy consumption from the database. Determination result data indicating the determination result by determining whether or not to perform daily life activity including at least one of sleep, cooking and bathing by the resident of the home by acquiring the amount data and performing calculation processing A calculation unit that outputs a time period when the calculation unit detects a period having characteristics at the time of execution of the daily life behavior for the time-series energy consumption indicated by the energy consumption data. It is determined that the daily life behavior is executed.

  According to this behavior determination system, it is possible to output determination result data that can grasp not only the presence / absence of a resident, but also the specific behavior of the resident, such as whether or not daily activities are performed. Become.

  Further, in the behavior determination system having the above characteristics, the calculation unit normalizes the time-series energy consumption indicated by the energy consumption data using a maximum value and a minimum value of the energy consumption during a predetermined period. The arithmetic unit detects a period having characteristics at the time of execution of the daily life behavior for the energy consumption in the normalized time series, and determines that the daily life behavior is executed during the period. preferable.

  According to this behavior determination system, it is possible to suppress the influence of the energy consumption that is constantly generated and to increase the accuracy of the determination.

  Furthermore, in the behavior determination system having the above characteristics, sleep is included in the daily life behavior, and the calculation unit is a sleep time zone that is a predetermined time zone, and the energy consumption is a threshold value for sleep detection. If it is detected that the period is below, it is preferable to determine that sleep is performed during the period.

  According to this behavior determination system, the period during which the resident is sleeping can be determined separately from the period when the resident is absent.

  Furthermore, in the behavior determination system according to the above feature, the time when the sleep time period starts is the start time, the time when the sleep time period ends is the end time, and the predetermined time between the start time and the end time is the boundary time When the calculation unit, the energy consumption after the start time and before the boundary time, from the time when the energy consumption falls below the sleep detection threshold, after the boundary time and before the end time, It is preferable to determine that sleep has been performed in a period until the time when the amount exceeds the sleep detection threshold for the first time.

  According to this behavior determination system, it is possible to determine the period in which the resident is sleeping in the sleep time period, by distinguishing it from the period in which the resident is absent in the sleep time period. .

  Furthermore, in the behavior determination system having the above characteristics, when the calculation unit detects a period in which the energy consumption amount continues for a predetermined time or more and falls below a sleep detection threshold in the sleep time period, the period It is preferable to determine that sleep has been performed.

  According to this behavior determination system, it is possible to suppress a period in which the activity level has temporarily decreased due to some factor in the sleep time period as a period in which sleep has been performed.

  Furthermore, in the behavior determination system of the above feature, when the time zone other than the sleep time zone is an activity time zone, the calculation unit is the activity time zone immediately before the sleep time zone and the sleep time zone, When it is detected that the energy consumption amount does not exceed the sleep detection threshold, it is preferable to determine that sleep has not been executed during the sleep time period.

  According to this behavior determination system, since the activity level is continuously low before and after the sleep time zone, it is suppressed that the sleep is executed when it is estimated that the resident is not going home. It becomes possible to do.

  Furthermore, in the behavior determination system having the above characteristics, each of the energy consumption data includes a power consumption that is an amount of power consumed at the home and a gas consumption that is a gas amount that is consumed at the home. The temporal variation is separately shown, and the calculation unit includes: the time-series power consumption indicated by the energy consumption data; and the time-series gas consumption indicated by the energy consumption data. For each of them, it is preferable to detect that an overlapping period having characteristics at the time of execution of the daily life action is determined to have executed the daily life action during the period.

  According to this behavior determination system, determination can be made based on a plurality of energy consumptions such as power consumption and gas consumption, so that multifaceted determination can be performed, so that determination accuracy can be improved. .

  Further, in the behavior determination system having the above characteristics, the calculation unit normalizes the time-series power consumption indicated by the energy consumption data using a maximum value and a minimum value of the power consumption in a predetermined period. The computing unit detects, for each of the normalized power consumption and the gas consumption, an overlapping period having characteristics at the time of execution of the daily life behavior, It is preferable to determine that the daily life behavior has been executed.

  According to this behavior determination system, normalize power consumption that has a large impact on determination accuracy because it can occur constantly, and do not normalize gas consumption that is unlikely to occur constantly and has a small impact on determination accuracy. This makes it possible to avoid unnecessary arithmetic processing.

  Furthermore, in the behavior determination system having the above characteristics, cooking is included in the daily life behavior, and the calculation unit is configured such that the power consumption exceeds a cooking detection power amount threshold, and the gas consumption is detected as cooking. It is preferable to detect that cooking has been performed during the period when the period exceeding the working gas amount threshold is detected.

  Further, in the behavior determination system of the above feature, the daily life behavior includes bathing, and the calculation unit is configured such that the power consumption exceeds a bathing detection power amount threshold and the gas consumption is bathing detection. When a period exceeding the working gas amount threshold is detected, it is preferable to determine that bathing has been performed during the period.

  The security system of the present invention includes the behavior determination system described above and a monitoring control unit that controls the monitoring state of the home based on determination result data output from the behavior determination system. To do.

  In addition, the resident monitoring system of the present invention includes the behavior determination system described above and an anomaly detection unit that detects the occupant's anomaly based on determination result data output by the behavior determination system. Features.

  According to the behavior determination system of the above feature, since the determination result data indicating the presence or absence of daily living behavior is output, data having high utility value is output to various systems that perform operations related to the state and behavior of the resident At the same time, it is possible to output data having high utility value as statistical data.

The block diagram shown about an example of a structure of the action determination system which concerns on embodiment of this invention. The graph shown about an example of the calculation method of an activity degree. The graph shown about an example of the determination method of sleep. The graph shown about an example of the determination method of cooking. The graph shown about an example of the determination method of bathing.

<< Configuration of action determination system >>
First, an example of the configuration of the behavior determination system according to the embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a block diagram showing an example of the configuration of the behavior determination system according to the embodiment of the present invention.

  As shown in FIG. 1, the behavior determination system 1 according to the embodiment of the present invention includes a database 11 and a calculation unit 12. The database 11 is configured to include a recording device such as a hard disk or a semiconductor memory. Moreover, the calculating part 12 is the structure provided with arithmetic devices, such as CPU (Central Processing Unit), and the memory which memorize | stores data temporarily at the time of a calculation, for example.

  The database 11 records energy consumption data. The energy consumption data is data indicating temporal variation of energy consumption, which is the amount of energy consumed at home. In the behavior determination system 1 illustrated in FIG. 1, energy consumption data is created by an energy consumption measuring device 20 (for example, a smart meter) installed in a home.

  The energy consumption measuring device 20 indicates time-series energy consumption by recording the amount of power and gas consumed at unit time (for example, every few minutes, every few tens of minutes, etc.) at any time at home. Create energy consumption data. Then, the energy consumption measuring instrument 20 stores energy consumption data with respect to the database 11 at a predetermined timing (for example, several times a day, once every few days, etc. every time an electric power amount or a gas amount is recorded). Enter. However, the timing at which the energy consumption meter 20 records the amount of power and gas and the timing at which the energy consumption meter 20 inputs energy consumption data to the database 11 are determined by the behavior determination system 1. It is assumed that the request of the determination result data use unit 30 that uses the result data is met.

  The calculation unit 12 obtains energy consumption data from the database 11 and performs calculation processing to determine the behavior of a resident at home, and outputs determination result data indicating the determination result. In particular, the calculation unit 12 outputs determination result data indicating whether or not the daily living behavior is performed by the resident based on the energy consumption data.

  Here, the daily life behavior in which the calculation unit 12 determines whether or not to execute is, for example, an action that is inevitably and repeatedly executed by a normal person in daily life, and typically includes “sleep”, "Cooking" and "bathing" are mentioned. In the following, a case where the daily life behavior in which the calculation unit 12 determines whether or not to execute includes all of “sleeping”, “cooking”, and “bathing” as typical examples will be described. If even one is included, the determination result data output by the calculation unit 12 is data having high utility value as data indicating whether or not daily activities are performed (details will be described later).

  The determination result data output by the behavior determination system 1 (particularly, the calculation unit 12) can be used for, for example, a security system or a resident monitoring system. The determination result data utilization unit 30 shown in FIG. 1 is an element that constitutes a part of the system (monitoring control unit, anomaly detection unit) that uses the determination result data, such as a security system or a resident monitoring system.

  As described above, the determination result data output by the action determination system 1 indicates whether or not the resident is performing daily life actions. Therefore, in the system that uses the determination result data (the system including the action determination system 1 and the determination result data utilization unit 30 in FIG. 1), the residence is not limited to the presence / absence of the resident, but the presence / absence of the execution of daily life behavior. It is also possible to grasp the specific action content of the person. Therefore, in the system using this determination result data, it is possible to perform an appropriate operation according to the resident's state and behavior.

  For example, in a security system that uses this determination result data, the determination result data utilization unit 30 (monitoring control unit) that controls the monitoring state of the home based on the determination result data is in a state in which the resident who is at home is unprotected ( For example, it can be determined whether the user is sleeping or taking a bath. Therefore, if the occupant is in an unprotected state, the determination result data utilization unit 30 increases the number of sensors that are operated compared to the case in which the resident is not in an unprotected state (for example, activates a sensor that detects opening and closing of doors and windows). ) And the like, it is possible to realize an appropriate monitoring state according to the resident's state.

  Further, for example, in the resident monitoring system that uses the determination result data, the determination result data use unit 30 (anomaly detection unit) that detects an occupant's anomaly based on the determination result data causes the resident to perform daily life behavior It is possible to quickly grasp that it has stopped. Therefore, in this resident monitoring system, it is possible to quickly detect that a change has occurred in the resident (and the specific content of the change estimated from the daily life behavior that is no longer performed). Become.

  As described above, the behavior determination system 1 according to the embodiment of the present invention outputs determination result data indicating the presence / absence of daily living behavior, and therefore, for various systems that perform operations related to the resident's state and behavior. It is possible to output data with high utility value.

  Note that the above-described security system and resident monitoring system are merely examples, and the determination result data output by the behavior determination system 1 should be preferably used in various systems that perform operations related to the resident's state and behavior. Is possible.

  Further, the determination result data output by the behavior determination system 1 itself has high utility value as statistical data. Therefore, the determination result data output by the behavior determination system 1 is preferably used in, for example, a database system that accumulates determination result data (a system in which the determination result data use unit 30 is a recording device such as a hard disk or a semiconductor memory). Is possible.

  Moreover, although FIG. 1 illustrates the case where there is one energy consumption measuring instrument 20 that inputs energy consumption data to the database 11, a plurality of energy consumption measuring instruments 20 exist. Also good. For example, when the behavior determination system 1 determines each of the behaviors of residents in a plurality of homes, the energy consumption measuring device 20 installed in each home is used for the database 11 to store the energy consumption of each home. Quantity data may be entered.

  Further, FIG. 1 illustrates the case where the energy consumption measuring instrument 20 inputs energy consumption data to the database 11, but if the database 11 can record the energy consumption data, the database 11 may be used to input energy consumption data. For example, a meter reader connects a portable terminal or a recording device to the energy consumption measuring instrument 20 to acquire energy consumption data, and connects the portable terminal or the recording device to the database 11. The energy consumption data may be input to the database 11.

<< Operation of action determination system >>
Next, operation | movement of the action determination system 1 which concerns on embodiment of this invention is demonstrated with reference to drawings. In particular, a method for determining whether or not the daily living behavior is performed by the calculation unit 12 will be described below with a specific example with reference to the drawings. Moreover, below, the case where the action determination system 1 determines the action of the resident of the home where the heat source of a cooking appliance and a hot-water supply apparatus is gas and the energy source of an air-conditioning apparatus and another apparatus is electric power is demonstrated. .

<Calculation of activity>
First, the “activity level” that is referred to when the calculation unit 12 determines whether or not daily activities are performed will be described with reference to FIG. 2. FIG. 2 is a graph showing an example of an activity level calculation method. FIG. 2 (a) is a graph showing temporal fluctuations in the power consumption of a home on a certain day (from 0:00 to 24:00), and FIG. 2 (b) is shown in FIG. 2 (a). It is a graph which shows the activity computed from electric power consumption.

The calculation unit 12 normalizes the time-series power consumption shown in FIG. 2A by using the maximum value C _MAX and the minimum value C _MIN of the power consumption in a predetermined period, so that FIG. ) Is calculated.

First, the calculation unit 12 obtains the maximum value C _MAX and the minimum value C _MIN with reference to a predetermined period in which the pattern of power consumption is estimated to be similar to the period for obtaining the activity level. For example, the calculation unit 12 may have a predetermined period before and after a period for determining the activity level (for example, several tens of days before and after), a predetermined period before the period for determining the activity level (for example, the last several tens of days), or an activity The maximum value C _MAX and the minimum value C _MIN are determined with reference to a predetermined period in the past (for example, several tens of days before one year) that is the same period as the period for determining the degree.

For example, when it is necessary to determine the resident's behavior in real time, as in the security system and the monitoring system described above, the power consumption after the period for obtaining the activity level is unknown. Therefore, in this case, the maximum value C _MAX and the minimum value C _MIN may be obtained from a predetermined period before the period for determining the activity level or a past predetermined period that is the same period as the period for determining the activity level.

Further, for example, when the determination result data is accumulated for use as statistical data, and the power consumption after the period for determining the activity is known, a predetermined period before the period for determining the activity is determined. The maximum value C _MAX and the minimum value C _MIN may be obtained from a predetermined period in the past, which is the same period as the period for _obtaining the activity level, or the maximum from the predetermined period before and after including the period for obtaining the activity level. The value C _MAX and the minimum value C _MIN may be obtained.

Next, the calculating part 12 calculates an activity degree by following formula (1), for example. In Formula (1), X _T is the power consumption at any time-of-day information T, the F _T is activity at the time T.

F _T = (X _T -C _MIN ) / (C _MAX -C _MIN ) (1)

  As shown in FIG. 2 (a), it is always constant at home regardless of the presence / absence of the resident and the content of the action due to the influence of the power consumption of a constantly operating device such as a refrigerator and the standby power of various devices. A large amount of power is consumed. The constantly generated power consumption does not vary so much in the short term, but may vary in the long term. For this reason, when the calculation unit 12 directly determines the presence / absence of execution of daily living behavior by directly using the power consumption, the determination result varies depending on the time, and the determination accuracy may decrease.

Thus, the calculation unit 12 determines the power consumption based on the activity level normalized using the maximum value C _MAX and the minimum value C _MIN , thereby suppressing the influence of the power consumption that is constantly generated, It becomes possible to improve the accuracy of determination.

  The activity level is only a normalized value of power consumption, which is a kind of energy consumption. Therefore, the calculation unit 12 determining whether or not the daily life behavior is performed based on the activity level is an aspect in which the calculation unit 12 determines whether or not the daily life behavior is performed based on the energy consumption. To be interpreted.

<Determination of sleep>
Next, an example of a method for determining whether or not “sleep”, which is one of daily activities, is performed by the calculation unit 12 will be described with reference to the drawings. FIG. 3 is a graph illustrating an example of a sleep determination method. For the sake of space, FIG. 3 shows only the activity level for one day (from 0:00 to 24:00), but it is assumed that the activity level of the same pattern is repeated every day.

  As shown in FIG. 3, in the sleep determination method of this example, the calculation unit 12 determines whether or not sleep is performed based on the activity level. First, in the sleep determination method of this example, in the sleep time zone (in the example of FIG. 3 from 19:00 to 13:00 on the next day) that is assumed to be a resident's sleep, in other time zones A certain activity time zone (from 13:00 to 18:00 in the example of FIG. 3) is set.

  For example, as the sleeping hours, it is estimated that the sleeping hours will be executed based on the sleeping hours when the resident declares normal times, or family information (eg, the presence or absence of infants and children, occupations, etc.) declared by the resident. It is possible to apply a belt, a night time zone in which a normal person is supposed to sleep, and the like. In addition, for example, a time period calculated based on the characteristics of the energy consumption amount (for example, activity level) indicated by the energy consumption amount data can be applied as the sleep time period. Specifically, for example, it is possible to apply a time period in which the activity level of a plurality of days is compared and a time period in which the activity level tends to decrease is recognized as the sleep time period.

In the determination method of sleep in this example, it determines that the operation unit 12, a band sleeping hours, when the activity is detected a period below the sleep threshold value for detecting A _1, sleep during the period has been executed. In this case, the arithmetic unit 12, sleep activity a time zone determines that sleep is performed in the period is below the threshold A ₁ for detecting sleep, an active time period is other time periods determines that the period during which the activity is below the threshold value a ₁ for detecting sleep not running sleep (resident is absent).

Furthermore, in the sleep determination method of this example, the time when the sleep time zone starts (19:00 in the example of FIG. 3) is the start time T _S , and the time when the sleep time zone ends (13:00 in the example of FIG. 3). A predetermined time (3 o'clock in the example of FIG. 3) between the end time T _E , the start time T _S and the end time T _E is set as the boundary time T _D. Then, computing unit 12, the start time T _S after and from the time boundary time T _D before activity is below the threshold value A ₁ for detecting sleep, boundary time T _D after and activity a end time T _E before There determines the period until the time exceeds the first time sleep detection threshold a _1, and sleep is executed.

  In the sleep determination method of this example, by setting and determining the sleep time zone and the activity time zone, the period during which the resident is sleeping (in the example of FIG. 3 from 22:00 to 6 o'clock the next day) It is possible to make a determination by distinguishing from a period in which the resident is absent (in the example of FIG. 3 from 14:00 to 17:00).

Furthermore, in the sleep determination method of this example, the start time T _S , the boundary time T _D, and the end time _TE are set and determined in the sleep time zone, so that the resident performs sleep in the sleep time zone. The period (22:00 in the example of FIG. 3 to 6 o'clock the next day) is distinguished from the period in which the resident is absent (8 to 12 in the example of FIG. 3). It becomes possible.

Incidentally, continued operation unit 12, the band sleep time (in particular, the start time T _S after and boundary time T _D previously), when determining the execution of the sleep, activity a predetermined time (e.g., several hours) or more if detecting a period in which below the sleep threshold value for detecting a ₁ and may not be determined that the sleep is executed.

  If it determines in this way, it will become possible to suppress determining the period when the activity level fell temporarily for some reason as a period when sleep was performed in the sleep time zone.

The arithmetic unit 12, an activity time zone of the immediately preceding sleep time period and the sleep time period, if the activity is detected that does not exceed the sleep detection threshold A _1, sleep on the sleeping time zone performed It may be determined that the resident is absent.

  If determined in this way, the activity level is continuously low before and after the sleep time period, so that it is suppressed that the sleep is determined to be performed when the resident is estimated not to go home. It becomes possible.

<Judgment of cooking>
Next, an example of a method for determining whether or not “cooking”, which is one of daily activities, is performed by the calculation unit 12 will be described with reference to the drawings. FIG. 4 is a graph illustrating an example of a cooking determination method. FIG. 4A is a graph showing temporal changes in activity from 16:00 to 24:00, and FIG. 4B is a graph showing gas consumption from 16:00 to 24:00.

  As shown in FIG. 4, in the cooking determination method of this example, the calculation unit 12 determines whether cooking is performed based on both the activity level and the gas consumption. In particular, the calculation unit 12 determines that cooking has been performed for each of the activity level and the gas consumption amount during the overlapping period having characteristics at the time of cooking.

Specifically, the calculation unit 12 is a period in which the activity level exceeds the cooking detection power amount threshold A ₂ and the gas consumption exceeds the cooking detection gas amount threshold G ₂ (in the example of FIG. 4 from 18:00 to 19 When (time) is detected, it is determined that cooking has been performed during the period.

In the example shown in FIG. 4, when a resident performs cooking, equipment such as a ventilation fan, lighting, a microwave oven, and an oven is used, and an increase in power consumption is assumed due to opening and closing of the refrigerator. because it is, has a cooking detection power amount threshold a ₂ greater than the sleep detection threshold a _1. Further, in the example shown in FIG. 4, a slight because it is assumed that cooking with gas consumption, is set to 0 m ³ cooking detecting gas amount threshold G _2.

<Judgment of bathing>
Next, an example of a method for determining whether or not “bathing”, which is one of daily activities, is performed by the calculation unit 12 will be described with reference to the drawings. FIG. 5 is a graph showing an example of a method for determining bathing. FIG. 5A is a graph showing temporal variation in activity from 16:00 to 24:00, and FIG. 5B is a graph showing gas consumption from 16:00 to 24:00.

  As shown in FIG. 5, the determination method for bathing in this example is the same as the example of the determination method for cooking described above. That is, the calculation unit 12 determines whether or not bathing is performed based on both the activity level and the gas consumption. In particular, the calculation unit 12 determines that the bathing has been performed for each of the activity level and the gas consumption amount during the overlapping period having characteristics at the time of performing the bathing.

Specifically, the calculation unit 12 is a period in which the activity exceeds the bathing detection power amount threshold A ₃ and the gas consumption exceeds the bathing detection gas amount threshold G ₃ (in the example of FIG. 5, from 20:00 to 21 When (time) is detected, it is determined that bathing was performed during the period.

In the example shown in FIG. 5, when the resident performs bathing, it is assumed that power consumption is reduced by stopping lighting and other devices other than the bathroom. the a ₃ are the same size as the threshold value a ₁ for sleep detection. Further, in the example shown in FIG. 5, since it is assumed that the gas consumption amount becomes large at the time of bathing, the bathing detection gas amount threshold value G ₃ is set to a value larger than 0 m ³ .

Further, as in the respective examples shown in FIGS. 4 and 5, the cooking detection power amount threshold A ₂ is set to a value larger than the bathing detection power amount threshold A ₃ , and the cooking detection gas amount threshold G ₂ is set to be larger. by a value smaller than a bath detecting gas amount threshold G _3, operation section 12, it is possible to determine distinguishes between bathing and cooking.

<Deformation, etc.>
[1] The sleep, cooking, and bathing determination methods illustrated in FIGS. 3 to 5 are based on normalized power consumption (activity), but the power consumption is not normalized. The determination may be performed based on the determination. However, as described above, the determination accuracy of the calculation unit 12 can be improved by performing the determination based on the normalized power consumption.

[2] The determination methods of cooking and bathing illustrated in FIGS. 4 and 5 are based on the power consumption and the gas consumption, but the power consumption (activity that normalized the power consumption) Determination may be made based only on However, by making a determination based on a plurality of energy consumption amounts such as the power consumption amount and the gas consumption amount, a multifaceted determination can be made, so that the determination accuracy can be improved.

  Moreover, the determination method of cooking and bathing illustrated in FIG.4 and FIG.5 has the characteristic at the time of execution of cooking or bathing in each of power consumption (activity) and gas consumption, and the period which overlaps. Only to determine that cooking or bathing has been performed. However, in addition to this, it may be determined that cooking or bathing has been performed for a period having characteristics at the time of performing cooking or bathing detected from one of power consumption and gas consumption.

  However, as described above, if it is determined that cooking or bathing has been performed only during the overlapping period with characteristics at the time of cooking or bathing, it is determined by accurately distinguishing whether or not cooking or bathing has been performed. It becomes possible to do.

  Moreover, although the determination method of cooking and bathing illustrated in FIG.4 and FIG.5 performs determination based on the normalized power consumption (activity) and the gas consumption which is not normalized, it normalizes The determination may be made based on the power consumption and the normalized gas consumption.

  However, since it is difficult to assume the consumption of gas at the time of standby of the gas equipment or the gas equipment at all times, unlike the power consumption, the effect of normalizing the gas consumption is small. Therefore, it is preferable to normalize the power consumption and not to normalize the gas consumption as in the cooking and bathing determination methods illustrated in FIGS. 4 and 5, because unnecessary calculation processing can be avoided. .

[3] The activity calculation method illustrated in FIG. 2 calculates the activity based on one type of energy consumption (power consumption), but energy consumption is a combination of multiple types of energy consumption. The degree of activity may be calculated based on the amount (for example, a weighted addition value of power consumption and gas consumption). In this case, the calculation unit 12 may calculate the activity level by normalizing the combined energy consumption by the same method as the calculation method illustrated in FIG.

  Moreover, the calculating part 12 may determine a resident's action based on the energy consumption after the synthesis | combination which is not normalized.

[4] In the above-described embodiment, it is assumed that the heat source for cooking and hot water supply is gas. However, even if the heat source for cooking and hot water supply is electric power (so-called all-electric housing), behavior determination is performed. System 1 is available. In this case, as illustrated in FIG. 4 and FIG. 5, in order to determine whether or not cooking and bathing are performed, the behavior determination system 1 uses power consumption (or activity) instead of gas consumption. What is necessary is just to detect the period which has the characteristic (For example, fluctuation | variation of the power consumption accompanying use of an electric cooker or an electric water heater) at the time of execution of cooking and bathing.

[5] When a period having a characteristic at the time of execution of a certain daily life action also has a characteristic at the time of execution of another daily life action, the behavior determination system 1 has a priority set in advance for the period. It may be determined that a high daily life behavior has been executed.

  Specifically, for example, when higher priority is set for bathing than cooking, when a period having both characteristics at the time of cooking and characteristics at the time of bathing is detected, The behavior determination system 1 may determine that bathing has been performed during the period.

[6] In the above-described embodiment, the energy consumption data has been described as separately indicating each of the temporal fluctuation of the power consumption and the temporal fluctuation of the gas consumption. It is also possible to show only the fluctuations in time of consumption of energy and resources of different types (for example, consumption of water supply).

  INDUSTRIAL APPLICABILITY The present invention can be used for a behavior determination system that determines the behavior of a resident, a security system that uses a determination result of the behavior determination system, and a resident watch system.

1: Action determination system 11: Database 12: Calculation unit 20: Energy consumption measuring device 30: Determination result data utilization unit (monitoring control unit, anomaly detection unit)

Claims (9)

A database that records energy consumption data indicating temporal fluctuations in energy consumption, which is the amount of energy consumed at home;
By obtaining the energy consumption data from the database and performing arithmetic processing, it is determined whether or not daily living activities including sleep, cooking, and bathing are performed by residents of the home, A calculation unit that outputs determination result data indicating the determination result, and
The energy consumption data separately indicates temporal variations of power consumption, which is the amount of power consumed at the home, and gas consumption, which is the amount of gas consumed at the home. Yes,
The calculation unit normalizes the time-series power consumption indicated by the energy consumption data using a maximum value and a minimum value of the power consumption in a predetermined period, and calculates an activity level.
When the computing unit detects a period having a characteristic at the time of execution of the daily life action for the daily activity including at least sleep , the daily life action is detected during the period. Determined that it was executed
The calculation unit is configured for each of the activity in time series and the gas consumption in time series indicated by the energy consumption data with respect to daily activities including at least one of cooking and bathing. When the period which has the characteristic at the time of execution of the said everyday life action is detected and it overlaps, it determines with the said everyday life action being performed in the said period .   When the arithmetic unit detects a period in which the activity level is lower than a threshold for sleep detection in a sleep period that is a predetermined period, it is determined that sleep is performed in the period. The behavior determination system according to claim 1. When the time when the sleep time period starts is the start time, the time when the sleep time period ends is the end time, and the predetermined time between the start time and the end time is the boundary time,
The calculation unit is configured to start from the time when the activity falls below the sleep detection threshold after the start time and before the boundary time, and after the boundary time and before the end time, and the activity level is for the sleep detection. The behavior determination system according to claim 2, wherein sleep is determined to be executed during a period until a time when the threshold value is first exceeded.   When the calculation unit detects a period in which the activity level continues for a predetermined time or more and falls below a sleep detection threshold in the sleep time period, the calculation unit determines that sleep is performed during the period. The behavior determination system according to claim 2 or 3. When a time zone other than the sleep time zone is an activity time zone,
When the calculation unit detects that the activity level does not exceed the sleep detection threshold in the sleep time zone and the activity time zone immediately before the sleep time zone, sleep is executed in the sleep time zone. It determines with having not existed, The action determination system of any one of Claims 2-4 characterized by the above-mentioned. Cooking is included in the daily life behavior,
If the said activity part detects the period when the said activity level exceeds the electric energy threshold value for cooking detection, and the said gas consumption exceeds the gas amount threshold value for cooking detection, it will determine with cooking having been performed in the said period. The behavior determination system according to claim 1, wherein the behavior determination system is characterized in that The daily activities include bathing,
The arithmetic unit, when detecting a period in which the activity level exceeds a bathing detection power amount threshold and the gas consumption exceeds a bathing detection gas amount threshold, determines that bathing has been performed during the period. The behavior determination system according to claim 1, wherein the behavior determination system is characterized in that The behavior determination system according to any one of claims 1 to 7 ,
Based on the determination result data output by the behavior determination system, a monitoring control unit that controls the monitoring state of the home;
A security system characterized by comprising: The behavior determination system according to any one of claims 1 to 7 ,
Based on the determination result data output by the behavior determination system, an anomaly detection unit that detects an anomaly of the resident,
Resident watch system characterized by having.

Images