JP5985446B2 - Life situation estimation method, estimation apparatus, and estimation program - Google Patents

Description

  The present invention relates to a living state estimation method, an estimation apparatus, and an estimation program. More specifically, the present invention relates to a technique suitable for use in confirming the safety of a resident in a single-person household of an elderly person or a disabled person or a household composed of only an elderly person and a disabled person.

  As a conventional system for confirming the safety of a resident, for example, a plurality of infrared sensors that are provided in various places including a predetermined room where a bed is arranged in a safety confirmation target person's house and are sensitive to infrared rays from the human body, In the life monitoring system comprising the presence determination means for determining the location of the human body based on the sensitivity signal of the infrared sensor, that is, the whereabouts of the target person, the living room monitoring system further includes a bed sensor sensitive to the human body only in the bed, When there is a response of the bed sensor, it is determined whether or not a human body is present on the bed based only on the bed sensor.

JP 2003-317165 A

  In the life monitoring system of Patent Document 1, it is necessary to provide a plurality of infrared sensors at various locations of the safety confirmation target person's house, resulting in high costs. In addition, providing an infrared sensor at each location of the safety confirmation subject's home and determining the resident's whereabouts based on the sensitivity signal of the infrared sensor means that the details of the living situation for the residents of the safety confirmation subject's home are There is a problem that it is not comfortable because it is always monitored by a third party and the lifestyle is known, and there is a possibility that it may give the resident discomfort.

  Therefore, the present invention does not provide an additional function or device in the household, and does not require information on a specific living situation such as the location of the resident in the household, for example, confirmation of safety of the resident. An object of the present invention is to provide a living situation estimation method, an estimation apparatus, and an estimation program that can estimate a living situation that can be performed.

  In order to achieve such an object, the living state estimation method of the present invention includes a step of measuring the total load current in the whole household by the measuring means, and a change in the total load current using the measurement data of the total load current. A step of calculating an absolute value and an X percentile value (X value is preset) in an ascending or descending order of absolute values of changes in a plurality of total load currents in the most recent predetermined time range are extracted. A step of setting the X percentile value of the current change absolute value, a step of calculating an integrated value of the X percentile values of the current change absolute value in a preset time zone by dividing the time zone of the day, and a current change Using the integrated value of the X percentile value of the absolute value, the average value of the integrated value of the X percentile value of the absolute value of the current change in the time zone in the most recent predetermined period range The step of calculating the standard deviation and the integrated value of the X percentile value of the current change absolute value and the average value of the integrated value of the X percentile value of the current change absolute value are the integrated value of the X percentile value of the current change absolute value And a step of estimating the living situation of the resident of the household by taking into account the standard deviation.

  Further, according to the method for estimating the living situation of the present invention, the step of measuring the total load current in the whole household by the measuring means, and the absolute value of the change of the total load current using the measurement data of the total load current is The calculated step and the X percentile value (X value is preset) in the ascending order or descending order of the absolute values of the plurality of changes in the total load current in the most recent predetermined time range are extracted to change the current An absolute value X-percentile value step, a step of calculating an X-percentile value integrated value of the current change absolute value in the latest past predetermined range, and an integrated value of the X-percentile value of the current change absolute value are used. The average value and the standard deviation of the integrated values of the X percentile values of the current change absolute value in the most recent past predetermined range in the most recent past predetermined range. And the integrated value of the X percentile value of the current change absolute value and the average value of the X percentile value of the current change absolute value are the integrated value of the X percentile value of the current change absolute value. And a step of estimating the living situation of the resident of the household by taking into account the standard deviation.

  The life situation estimation apparatus according to the present invention also includes means for reading data on the total load current of the entire household measured by the measuring means from the storage device, and the absolute change in the total load current using the data on the total load current. A means for calculating a value, and extracting an X percentile value (X value is preset) in an ascending order or descending order of absolute values of changes in a plurality of total load currents in the last past predetermined time range, and a current change Means for calculating the Xth percentile value of the absolute value, means for calculating the integrated value of the Xth percentile value of the absolute value of current change in a preset time zone when the time zone of the day is divided, and X of the absolute value of current change A method for calculating an average value and a standard deviation of the integrated values of the X percentile values of the absolute value of the current change in the time zone in the most recent predetermined period range using the integrated values of the percentile values. And the integrated value of the X percentile value of the current change absolute value and the average value of the integrated value of the X percentile value of the current change absolute value, taking into account the standard deviation of the integrated value of the X percentile value of the current change absolute value And a means for estimating the living conditions of the resident of the household.

  The life situation estimation apparatus according to the present invention includes a means for reading data of a total load current of the entire household measured by the measuring means from a storage device, and a change in the total load current using the data of the total load current. And an X percentile value (X value is set in advance) in an ascending order or descending order of absolute values of a plurality of changes in the total load current in a range of the last past predetermined time. Means for calculating the X percentile value of the current change absolute value, means for calculating the integrated value of the X percentile value of the current change absolute value in the most recent predetermined range, and the integrated value of the X percentile value of the current change absolute value The average value and standard deviation of the integrated values of the X percentile values of the current change absolute value in the most recent past predetermined range in the range of the most recent past predetermined period are used. A standard deviation of the integrated value of the X-percentile value of the current change absolute value, and the average value of the integrated value of the X-percentile value of the current change absolute value and the integrated value of the X-percentile value of the current change absolute value And a means for estimating the living situation of the resident of the household.

  Further, the life situation estimation program of the present invention is a means for reading data on the total load current of the entire household measured by the measuring means from the storage device, and the absolute value of the change in the total load current using the data on the total load current. Calculating the absolute value of the current change by extracting an X percentile value (X value is preset) in an ascending order or descending order of the absolute values of a plurality of changes in the total load current in the last past predetermined time range Means for calculating the X percentile value of the current change absolute value, means for calculating the integrated value of the X percentile value of the absolute value of current change in a preset time zone when the time zone of the day is divided, integration of the X percentile value of the current change absolute value Means for calculating an average value and a standard deviation of the integrated values of the X percentile values of the absolute value of the current change in the time period in the most recent predetermined period range using the values Household by comparing the integrated value of the X percentile value of the current change absolute value and the average value of the integrated value of the X percentile value of the current change absolute value, taking into account the standard deviation of the integrated value of the X percentile value of the current change absolute value The computer is made to function as a means for estimating the living conditions of the residents.

  Further, the life situation estimation program of the present invention is a means for reading data on the total load current of the entire household measured by the measuring means from a storage device, and the change of the total load current using the data on the total load current. Means for calculating an absolute value, extracting an X percentile value (an X value is preset) in an ascending or descending order of absolute values of a plurality of changes in the total load current in the last past predetermined time range, Means for calculating the X percentile value of the absolute value of change, means for calculating the integrated value of the X percentile value of the absolute value of current change in the latest past predetermined range, and the past value using the integrated value of the X percentile value of the absolute value of current change An average value and a standard deviation of the integrated values of the X percentile values of the current change absolute value in the most recent predetermined range within the predetermined period range are calculated. The integrated value of the X percentile value of the current change absolute value and the average value of the integrated value of the X percentile value of the current change absolute value are taken into account with the standard deviation of the integrated value of the X percentile value of the current change absolute value. In contrast, the computer is made to function as a means for estimating the living situation of the resident of the household.

  Therefore, according to the estimation method, estimation device, and estimation program for the living situation, the average value of the integrated X-percentile values of the absolute value of the current change used when estimating the living situation of the resident of the household is It is set for each household based on the measured total load current. Therefore, the index value used when estimating the living situation of the resident of the household is set as a reflection of the characteristics (in other words, the life pattern) of the use of the electric equipment for each household. Moreover, an index value for each household is automatically set.

  According to the estimation method, estimation device, and estimation program of these living conditions, the measurement data of the total load current in the entire household is used, so that the specific living conditions such as the resident's indoor location can be used. Without the use of information, or more specifically, without the use of specific (in other words, individual) electrical equipment used in the household, the use of electrical equipment associated with indoor activities by residents. The presence or absence is judged and it is estimated whether the resident is living normally.

  According to these estimation methods, estimation devices, and estimation programs for living conditions, the living conditions of the resident can be confirmed at a place away from the household where the measuring means is installed, so that the resident's household is visited directly. It is possible to confirm whether or not the resident is living normally. In addition, since it is possible to check the status of multiple households on a single device, it is possible to check whether or not the residents of multiple households are living normally with little effort in normal operation. be able to.

  Further, according to the living state estimation method, estimation apparatus, and estimation program of the present invention, the time zone classification is from 3 am to 9 am, 9 am to 3 pm, 3 pm to 9 pm, 9 pm It may be set to four sections at 3:00 am on the next day. In this case, the living situation is estimated after each value is calculated and determined for each of the four sections.

  In addition, the living state estimation method, estimation apparatus, and estimation program of the present invention may be set within the range of the latest past predetermined time in the range of 15 to 60 minutes. In this case, the latest past predetermined time is set. Becomes appropriate.

  In addition, the life situation estimation method, estimation apparatus, and estimation program of the present invention may be set such that the most recent past predetermined period is set in a range of 5 to 30 days. In this case, the most recent past predetermined period is set. Becomes appropriate.

  In addition, the living state estimation method, estimation apparatus, and estimation program of the present invention may be set so that the most recent past predetermined range is 3 to 9 hours. In this case, the most recent past predetermined range is set. Becomes appropriate.

  According to the living situation estimation method, estimation apparatus, and estimation program of the present invention, the index value used to estimate the living situation of the resident of the household is reflected by the characteristics of electric appliance usage (in other words, the lifestyle pattern) for each household. Since it is possible to automatically set the measured value, it is not necessary to perform work that requires labor and experience of setting the index value by visual measurement etc. Even in the case of making it, it becomes possible to quickly set an appropriate index value for each household.

  According to the life situation estimation method, estimation apparatus, and estimation program of the present invention, the usage pattern of household electrical appliances changes according to seasonal changes, etc., and index values can be automatically set. Therefore, it is possible to constantly update the index value to reflect the change in the usage pattern of the electric equipment, and to always appropriately estimate the living situation of each household. In addition, by using the most recent measurement data when setting the index value, the index value can be made to reflect the household's most recent electrical equipment usage pattern, so appropriate estimation in line with the current situation of the household Can always be done.

  According to the estimation method, estimation apparatus, and estimation program of the living situation of the present invention, and without using the information on the concrete living situation such as the resident's indoor location, more specifically, the identification provided in the household Life of whether or not the resident is living steadily by judging whether or not the resident uses the electrical equipment associated with indoor activities without using the usage information (in other words, individual) of the resident Since the situation can be estimated, it is possible to prevent the residents from feeling uncomfortable, such as the details of the living situation being constantly monitored by a third party and the lifestyle being known. It is possible to alleviate obstacles to the privacy of system diffusion and promote the spread of the living situation estimation system, for example, to improve the living safety of single elderly households or households only for the elderly.

  According to the estimation method, estimation apparatus, and estimation program of the living situation of the present invention, it is possible to confirm the living situation of whether or not the resident is living normally without going directly to the resident's household. For example, it becomes possible to improve the living safety of a single household for the elderly and a household only for the elderly. In addition, since it is possible to check the living conditions of whether or not the residents of multiple households are living habitually with little effort in normal operation, it reduces the obstacles to the cost of disseminating the living situation estimation system. Thus, it is possible to promote the spread of the living situation estimation system, and to improve the living safety of, for example, a single elderly household or a household only for the elderly.

  In addition, each time period is divided into four times, from 3 am to 9 am, 9 am to 3 pm, 3 pm to 9 pm, and 9 pm to 3 am on the next day. If this is done for each of these categories, the life situation will be estimated after each value is calculated and determined for each category of the time zone of the day. The living situation can be estimated in consideration of the change, and further, for example, the household can be closely watched through the estimation of the living situation.

  In addition, when the most recent past predetermined time is set in the range of 15 to 60 minutes, the setting of the most recent past predetermined time can be made appropriate, and the estimation of the resident of the household is further improved. It becomes possible to do.

  In addition, when the most recent past predetermined period is set in the range of 5 to 30 days, the setting of the most recent past predetermined period can be made appropriate, and the estimation of the resident of the household is further improved. It becomes possible to do.

  Moreover, when the most recent past predetermined range is set in the range of 3 to 9 hours, the setting of the most recent past predetermined range can be made appropriate, and the estimation of the resident of the household is further improved. It becomes possible to do.

It is a flowchart explaining an example of embodiment of the estimation method of the living condition of this invention. It is a figure which shows an example of the whole structure of the apparatus put together by this invention. It is an example of the functional block diagram of the life condition estimation apparatus implement | achieved by the said program in the case of implementing the life condition estimation method of embodiment using the life condition estimation program. It is a flowchart explaining an example of other embodiment of the estimation method of the living condition of this invention. Shows the time course of the cumulative value M ₁₀ and alarm whether alarms and threshold "A ₁₀ -3Shiguma _10" of 10 percentile value of the current change absolute value in each time zone over a 60-Day certain household Example 1 FIG. It is the figure which extracted and showed the 56th-60th day of FIG.

  Hereinafter, the configuration of the present invention will be described in detail based on an example of an embodiment shown in the drawings.

  FIG. 1 to FIG. 3 show an example of an embodiment of the life situation estimation method, estimation apparatus, and estimation program of the present invention. Note that various electrical devices such as a television, a refrigerator, and a lighting fixture are connected to the indoor line in the household 9. And the indoor line of the household 9 is connected to electric power systems, such as an electric power company, through the lead-in wire which is not illustrated, and the electric wire laid by the utility pole.

  As shown in FIG. 1, the living state estimation method according to the present embodiment includes a step (S1) in which the measuring means 1 measures the total load current at the calculation time for the entire household 9, and the total load current measurement data. And calculating the absolute value of the change in the total load current (S2), and the X percentile value (X in the ascending order or descending order of the absolute values of the changes in the total load current in the most recent past predetermined time range. (S3) is extracted to obtain the Xth percentile value of the current change absolute value, and the calculation time is the last of the time zones set in advance by dividing the time zone of the day. Step (S4) in which it is determined whether or not it is time, step (S5) in which the integrated value of the X percentile value of the current change absolute value in the time zone is calculated, and the X parcel of the current change absolute value A step (S6) of calculating an average value and a standard deviation of the integrated values of the X percentiles of the current change absolute value in the time zone in the range of the latest past predetermined period using the integrated value of the tile value, and the current change absolute value The average value of the integrated value of the X percentile value of the current change absolute value and the average value of the integrated value of the X percentile value of the current change absolute value are compared with the standard deviation of the integrated value of the X percentile value of the current change absolute value. The living situation is estimated, specifically, there is a step (S7) for judging whether or not there is a possibility that an abnormal situation has occurred in the resident of the household 9.

  Moreover, the life situation estimation apparatus of the present embodiment includes means (11a) for reading data of the total load current at the calculation time of the entire household 9 measured by the measuring means 1 from the data server (2) as a storage device. A means (11b) for calculating the absolute value of the change in the total load current using the data of the total load current, and an ascending or descending order of the absolute values of the changes in the total load current in the most recent predetermined time range. Means (11c) for extracting the X percentile value (the value of X is preset) to obtain the X percentile value of the current change absolute value, and the current in the preset time zone by dividing the time zone of the day The means (11d) for calculating the integrated value of the X-percentile value of the change absolute value and the integrated value of the X-percentile value of the current change absolute value are used in the time zone within the most recent predetermined period. Means (11e) for calculating the average value and standard deviation of the integrated value of the X percentile value of the current change absolute value, the integrated value of the X percentile value of the current change absolute value, and the integrated value of the X percentile value of the current change absolute value (11f) for estimating the living situation of the resident of the household 9 by comparing the average value of the current value and the standard deviation of the integrated value of the X percentile value of the current change absolute value.

  Furthermore, the life situation estimation program of the present embodiment is a means (11a) for reading data of the total load current at the calculation time for the entire household 9 measured by the measuring means 1 from the data server (2) as a storage device, Means (11b) for calculating the absolute value of the change in the total load current using the data of the total load current, and the X percentile in the ascending order or descending order of the absolute values of the changes in the total load current in the most recent predetermined time range Means (11c) for extracting the value (the value of X is set in advance) to obtain the X percentile value of the current change absolute value, the current change absolute value in the time zone set in advance by dividing the time zone of the day Means (11d) for calculating the integrated value of the X percentile value of the current, and using the integrated value of the X percentile value of the absolute value of the current change in the time period within the most recent predetermined period Means (11e) for calculating the average value and standard deviation of the integrated value of the X percentile value of the current change absolute value, the integrated value of the X percentile value of the current change absolute value and the integrated value of the X percentile value of the current change absolute value The computer is caused to function as means (11f) for estimating the living situation of the resident of the household 9 by comparing the average value with the standard deviation of the integrated value of the X percentile value of the current change absolute value.

  In executing the method of estimating the living situation, first, the measuring means 1 measures the total load current in the entire household 9 and transmits measurement data (S1).

  The measuring means 1 is installed in the vicinity of the feeding line (leading line) inlet of the household 9, and measures the total load current (instantaneous value: unit [A]) at every predetermined time interval Δt. The measuring means 1 may measure total power consumption (instantaneous value: unit is [W]). The measuring means 1 is a device having a function of measuring the total load current (or total power consumption) of the entire household 9 and a function of outputting the measurement data of the total load current to the outside together with time information at a predetermined interval. If it exists, it is not limited to a specific device, and an existing device or a new device may be used.

  The value of the time interval Δt at which the measuring means 1 measures the total load current is not limited to a specific value, and the total load current of the entire household 9 accompanying the operation of the electrical equipment by the resident of the household 9 is not limited. A value that can detect the change, in other words, measurement is performed before and after the operation of the electrical device (specifically, switching on or off), and the change in the total load current caused by the operation is the time-series measurement data value. A value that can appear as a difference is appropriately selected. Specifically, for example, Δt can be appropriately selected and set within a range of about 1 to 5 [minutes] (however, it is not limited to 1 to 5 [minutes]). According to the study, about 1 minute is preferable.

  Note that the measuring means 1 may measure the integrated electric energy in the entire household 9 as an integrated value during the time interval Δt, not an instantaneous value. Specifically, for example, an existing watt-hour meter may be used as the measuring means 1.

  And the measuring means 1 measures the total load current in the whole household 9 for every time interval (DELTA) t, and measures the measurement data of the total load current acquired by the said measurement at a predetermined interval with time information. 2 to send. The time information includes information on measurement date and time.

  In the measured value data server 2, the time information and the total load current measurement data transmitted from the measuring means 1 are aggregated at the date d and time t for each household (in other words, associated with the identification information of the household). Accumulated as load current I (d, t) [A].

  Note that the communication means for data transmission between the measurement means 1 and the measurement value data server 2 is not limited to a specific mechanism, and the circumstances and measurement of the area or facility where the measurement means 1 is installed. An appropriate one is appropriately selected in consideration of the area where the value data server 2 is installed and the circumstances of the facility. Specifically, for example, use of an optical fiber, a metal line, a PLC (abbreviation of Power Line Communications: power line communication), a cellular phone line, a wired telephone line, a dedicated wireless line, and the like can be considered.

  Next, the absolute value ΔI (d, t) of the change of the total load current is calculated using the measurement data of the total load current measured by the process of S1 (S2).

  Here, the process after S2 in the living situation estimation method of the present invention can be executed by the living situation estimation apparatus of the present invention.

  And the process after S2 in the life situation estimation method of the present invention and the life situation estimation apparatus for executing these processes can also be realized by executing the life situation estimation program of the present invention on a computer. In the present specification, when the life situation estimation apparatus for executing the process after S2 is realized by executing the life situation estimation program on the computer, the process after S2 in the life situation estimation method is executed. Will be explained.

  FIG. 3 shows the overall configuration of a computer 10 (which is also a living situation estimation apparatus 10 in this embodiment) for executing the living situation estimation program 17. The computer 10 (living condition estimation device 10) includes a control unit 11, a storage unit 12, an input unit 13, a display unit 14, and a memory 15, and is connected to each other by a signal line such as a bus. Further, the computer 10 is connected to a measured value data server 2 as a storage device by a signal line such as a bus, and signals such as data and control commands are transmitted and received (ie, input / output) to each other via the signal line. Done.

  The control unit 11 performs calculation related to the control of the entire computer 10 and the estimation of the living situation by the living situation estimation program 17 stored in the storage unit 12, and is, for example, a CPU (Central Processing Unit).

  The storage unit 12 is a device that can store at least data and programs, and is, for example, a hard disk.

  The memory 15 serves as a memory space that is a work area when the control unit 11 executes various controls and operations, and is, for example, a RAM (abbreviation of Random Access Memory).

  The input unit 13 is an interface for giving at least an operator's command to the control unit 11, and is, for example, a keyboard.

  The display unit 14 performs drawing / display of characters, graphics, and the like under the control of the control unit 11 and is, for example, a display.

  In the present embodiment, the measurement data of the total load current measured in the above-described processing of S1 is stored (saved) in the measurement value data server 2 as the total load current database 18.

  Specifically, the combination data of the value of the total load current I (d, t) measured and transmitted by the measuring means 1 and the time information is associated with the identification information of the household 9 where the data is measured. And recorded in the total load current database 18.

  In addition, a calculation value data file 19 for appropriately recording (saving) various calculation values calculated in the processing to be described later is a storage unit 12 or a calculation value data server 2 or other appropriate storage device / storage medium. (Stored in the storage unit 12 in the example shown in FIG. 3).

  Then, the control unit 11 of the computer 10 (which is also the living situation estimation apparatus 10 in this embodiment) executes the living situation estimation program 17 so that the household measured by the measuring means 1 in the process of S1. 9 The data reading unit 11a that performs processing for reading the data of the total load current at the calculation time of the whole from the measured value data server 2 as a storage device, and the absolute value of the change in the total load current using the data of the total load current A change absolute value calculation unit 11b that performs a calculation process, and an X percentile value in an ascending order or descending order array of absolute values of changes in a plurality of total load currents in the latest past predetermined time range (the value of X is preset) Is extracted from the percentile value extraction unit 11c that performs processing to extract the current change absolute value as the X percentile value, and the time zone of the day A percentile value integrating unit 11d that performs processing for calculating an integrated value of the X percentile value of the current change absolute value in a preset time zone, and an integrated value of the X percentile value of the current change absolute value, and a predetermined past period An integrated value index calculating unit 11e that performs processing for calculating an average value and a standard deviation of the integrated values of the X percentile values of the current change absolute value in the time zone in the range, and an integrated value and current of the X percentile value of the current change absolute value The average value of the integrated value of the X percentile value of the change absolute value is compared with the standard deviation of the integrated value of the X percentile value of the current change absolute value, and a process of estimating the living situation of the resident of the household 9 is performed. The determination unit 11f is configured.

  As specific processing by executing the living situation estimation program 17, first, the total load current is measured by the data reading unit 11a configured in the control unit 11 of the computer 10 (living situation estimation device 10). Data is read (S2-1). In the following description, the process is described for one household 9. However, when there are a plurality of households to be processed, the same process is repeated for each household.

  Specifically, the measurement data (specifically, the total load current measured by the data reading unit 11a and recorded in the total load current database 18 stored (saved) in the measurement value data server 2 in the process of S1. Includes the total load current I (d, t) value and time information), the total load current measurement data at time t on date d, and the time [t before the time t [t Measurement data of the total load current of −Δt] is read.

  The read data of the total load current is stored in the memory 15 by the data reading unit 11a.

  Subsequently, the absolute value of the change in the total load current is calculated by the change absolute value calculation unit 11b of the control unit 11 based on the data of the total load current (S2-2).

Specifically, the change absolute value calculation unit 11b stores the value of the total load current I (d, t) at time t and the total load current at time [t−Δt] stored in the memory 15 in the process of S2-1. The value of I (d, t−Δt) is read from the memory 15, and the current change absolute value ΔI, which is the absolute value of the change in the total load current at a predetermined time interval Δt, is calculated by Equation 1.
(Equation 1) ΔI (d, t) = | I (d, t) −I (d, t−Δt) |

  In this embodiment, the value of time t of the later total load current among the two total load currents obtained as the difference is assigned to time t of the current change absolute value ΔI (d, t). Hereinafter, the time assigned to the current change absolute value ΔI (d, t) is referred to as a calculation time tc.

  In the following, the date of the current day of the calculation time tc that is the processing target in the ongoing processing step is denoted as date dc.

  Then, the calculated current change absolute value ΔI (dc, tc) is associated with the household identification information and combined with date d and time t by the change absolute value calculation unit 11b, and the calculated value data file 19 Is recorded (in other words, written).

  Here, measurement of the total load current of the household 9 by the measuring means 1 and transmission of the measurement data to the outside are automatically performed as a function of the measuring means 1. On the other hand, the processing from S2 is repeated every time interval Δt [minute], which is the measurement interval of the total load current by the measuring means 1. In the present embodiment, specifically, the process from S2 is triggered by the input of the measurement data of the total load current from the measurement means 1 to the measurement value data server 2 (in other words, recording in the total load current database 18). Done.

Next, the percentile value extraction unit 11c of the control unit 11 determines the X percentile value L _X of the current change absolute value in the range of the latest past predetermined time m [minutes] based on the current change absolute value ΔI (d, t). (S3).

  Specifically, the data of the current change absolute value (specifically, the current change absolute value ΔI (d, t) recorded in the calculation value data file 19 in the process of S2-2 by the percentile value extraction unit 11c. Current absolute value ΔI (d, t) in the range of the most recent predetermined time m [minutes] from the calculated time tc is read from the combination data of the value and date d / time t).

Then, the percentile value extraction unit 11c rearranges a plurality of current change absolute values ΔI (d, t) in the range of the latest past predetermined time m [minutes] in descending order, and extracts the X percentile values in the descending order array, The value is the X percentile value L _X of the current change absolute value.

The X percentile value L _X of the current change absolute value, which is the X percentile value in the descending order array, is, that is, “the current change absolute value in which the number of current change absolute values ΔI (d, t) that are greater than or equal to the value is arranged in descending order. Is a value that becomes X% with respect to the total number of values ΔI (d, t). Specifically, when the total number of current change absolute values ΔI (d, t) is n, “n × X / This is the value of the “100” th current change absolute value ΔI (d, t). If the value of “n × X / 100” is not an integer, the decimal part of the value of “n × X / 100” is rounded up, rounded down, or rounded off.

Here, the value of m [minute] which is the time as the range of the current change absolute value ΔI (d, t) used for determining the X percentile value L _X of the current change absolute value is limited to a specific value. It can be adjusted as appropriate. Specifically, for example, m can be appropriately adjusted and set within a range of about 15 to 60 [min] (however, it is not limited to 15 to 60 [min]). According to the study, about 30 minutes is preferable. Further, the time m [minutes] related to the determination of the X percentile value L _X of the current change absolute value is set to a value larger than the time interval Δt [minutes] for measuring the total load current of the household 9 by the measuring means 1 ( That is, m> Δt).

Further, the value of X [%] defining the percentile value extracted as the X percentile value L _X of the current change absolute value is not limited to a specific value and can be adjusted as appropriate. Note that a plurality of X [%] values may be set, and a plurality of X percentile values L _X of the current change absolute value may be extracted and used in the following processing. Specifically, for example, it is considered such that as used to extract the 10 percentile value L _10, 25 percentile value L _25, 50 percentile value L _50.

For example, the time interval Δt = 1 [minute] for measuring the total load current by the measuring means 1 and the range of the current change absolute value ΔI (d, t) used for determining the X percentile value L _X of the current change absolute value If the most recent predetermined time m = 30 [minutes], the number of current change absolute values ΔI (d, t) used to determine the X percentile value L _X of the current change absolute value is 30 [pieces]. is there.

Therefore, if X = 10 [%] that defines the percentile value, 30 × 10/100 = 3. Therefore, among the 30 current change absolute values ΔI (d, t) rearranged in descending order, the third largest current change value ΔI (d, t) is extracted as the 10th percentile value L _10. Is done. Similarly, the value of the eighth largest current change value ΔI (d, t) in the descending order is extracted as the 25th percentile value L ₂₅ , and the value of the fifteenth current change absolute value ΔI ( d, the value of t) is extracted as the 50 percentile value L _50.

Here, the X percentile value L _X of the current change absolute value is ΔI of the upper X% of the n current change absolute values ΔI measured during the most recent predetermined time m [minutes], and n ΔI that is equal to or greater than L _X among ΔI represents “n × X / 100”. By appropriately setting the value of _X , L _X is the total load current resulting from the operation of an electrical device such as a refrigerator that automatically repeats state transitions such as on / off regardless of resident operation. The change of the total load current caused by the operation of the electric device by the intentional operation of the resident of the household 9, in other words, the electric device by the resident of the household 9 in the most recent predetermined time m minutes It becomes an index that represents the amount of operation.

In the above description, a plurality of current change absolute values ΔI (d, t) are rearranged in descending order and the X percentile value is extracted. However, the present invention is not limited to this, and the most recent past predetermined time m A plurality of current change absolute values ΔI (d, t) in the range of [minute] are rearranged in ascending order, the X percentile value in the ascending order is extracted, and the value is set as the X percentile value L _X of the current change absolute value. You may make it.

  Here, in the present invention, the time zone of the day may be divided in order to accurately grasp the life pattern of the resident of the household 9. This time zone classification is based on the classification of the resident by calculating the index by dividing the time zone according to the classification and classifying the lifestyle (in other words, activity type) in the household of the household 9 It is intended to accurately capture lifestyle patterns. For example, focus on morning activities, daytime activities (including staying at home or going out), evening activities, and bedtime. It is possible to do. Therefore, the division of the time zone of the day in the present invention is not limited to a specific division method, and can be appropriately set in consideration of the rough lifestyle of the resident of the household 9.

Specifically, for example, it is conceivable to divide the time zone of the day into four sections every six hours as follows.
First time zone T ₁ : 3:00 am to 9:00 am Second time zone T ₂ : 9:00 am to 3:00 pm Third time zone T ₃ : 3:00 pm to 9:00 pm Fourth time zone T ₄ : Afternoon 9am-3am the next day

  In addition, the division | segmentation of the time zone of the day in this invention is not restricted to four divisions as said example, Two divisions, three divisions may be sufficient, and five divisions or more may be sufficient. In addition, the time for dividing each time zone is not limited to 3 am or 9 am in the above example. Furthermore, it is not necessary to equally divide the day, and the length may vary depending on the time zone. Furthermore, in the present invention, it is not essential to divide the time zone of the day, and the day time zone may be divided into units of one day, in other words, without distinction of the time zone. .

  The time zone is determined and determined based on the calculation time tc. In the following, a time zone including the calculation time tc that is the processing target in the ongoing processing step is referred to as a time zone Tc.

Based on the above, the X percentile value of the absolute value of current change in the time zone Tc of the date dc is expressed as L _X (dc, Tc, t). Note that the time t of the X percentile value L _X (dc, Tc, t) of the current change absolute value is the same as the calculation time tc determined in the process of S2-2.

In addition, the life style (in other words, activity type) in the household of the resident of the household 9 is greatly different between the weekday and the holiday, and the resident's life pattern is accurately calculated by calculating the index separately on the weekday and the holiday. If it can be grasped, the distinction between weekdays and holidays is further added to the classification of the X percentile value L _X of the current change absolute value (and the classification of each index calculated in the processing of S5 and S6 below). Also good.

Then, the current change absolute value ΔI (d, t) extracted as the X percentile value by the percentile value extraction unit 11c is used as the X percentile value L _X (dc, Tc, tc) of the current change absolute value. Is stored in the memory 15 as combined data of date d and time t.

  Next, the control unit 11 determines whether or not the calculated time tc is the final time of the time zone Tc. If the calculated time tc is not the final time of the time zone Tc (S4: No), the processing step is S1. When the calculation time tc is the final time of the time zone Tc (S4: Yes), the processing step proceeds to the processing of S5.

Specifically, for example, measurement of the total load current I of the household 9 by the measuring means 1 is performed every minute (that is, time interval Δt = 1 [minute]), and the calculation time tc to be processed is When belonging to the first time zone T ₁ (from 3 am to 9 am) in the above example, the start time of the time zone T ₁ is 3:00 am and the final time is 8 am It is 59 minutes.

If the calculated time tc is the final time of the time zone Tc (S4: Yes), the percentile value integrating unit 11d of the control unit 11 causes the X percentile value L _X (dc) of the current change absolute value in the time zone Tc. , Tc, t), an integrated value M _X (dc, Tc) is calculated (S5).

Specifically, the percentile value integrating unit 11d uses the X-percentile value data of the current change absolute value stored in the memory 15 in the process of S3 (specifically, the X-percentile value L _X (d , T, t) and date d / time t), the X percentile value L _X (dc, Tc, t) in the range of the time zone Tc is read from the memory 15.

Then, the percentile value accumulation unit 11d accumulates the X percentile value L _X (dc, Tc, t) of the current change absolute value in the range of the time zone Tc for each type of the X percentile value.

Specifically, for example, when three types of percentile values of the 10th percentile value L ₁₀ , 25th percentile value L ₂₅ , and 50th percentile value L ₅₀ of the current change absolute value are determined in the process of S3, The integrated values M ₁₀ , M ₂₅ , and M ₅₀ of the percentile values are calculated for every 10th percentile value, 25th percentile value, and 50th percentile value by 1-2-3.
(Equation 2-1) M ₁₀ (dc, Tc) = ΣL ₁₀ (dc, Tc, t)
(Equation 2-2) M ₂₅ (dc, Tc) = ΣL ₂₅ (dc, Tc, t)
(Equation 2-3) M ₅₀ (dc, Tc) = ΣL ₅₀ (dc, Tc, t)

Here, the integrated value M _X of X percentile value of the current change absolute value in the time zone Tc, during the time period Tc as an index representing an operation amount of the electric device by residents household 9 the most recent past predetermined time m min This is the total value of the extracted Xth percentile value L _X of the current change absolute value, and therefore represents the amount of activity (the total amount, corresponding to the integral value) for the electrical equipment by the resident of the household 9 in the time zone Tc. .

Then, the percentile value integrating section 11d associates the integrated value M _X (dc, Tc) of the X percentile value of the current change absolute value in the time zone Tc with the date d / time zone T in association with the household identification information. It is stored in the memory 15 as combination data.

Next, the integrated value index calculation unit 11e of the control unit 11, current change the absolute value of X percentile value of the integrated value M _X (dc, Tc) time zone in the region of the most recent past predetermined period k [day] based on Tc The average value A _X and the standard deviation σ _X of the integrated values of the X percentile values of the current change absolute value at are calculated (S6).

Specifically, the integrated value index calculation unit 11e uses the X-percentile value of the current change absolute value stored in the memory 15 in the processing of S5 up to the previous day (specifically, the X-percentile value integration). Among the values M _X (d, T) and the combined data of the date d and the time zone T) from the current day (ie, date dc) at the calculation time tc (ie, date [dc-1 ], [Dc-2], [dc-3],..., [Dc-k]), the integrated value M _X (d, Tc) of the X-percentile value of the current change absolute value in the time zone Tc. Is read.

Then, the integrated value index calculation unit 11e causes the average value A _X (dc) of the integrated values M _X (d, Tc) of the X percentile values of the current change absolute value in the time period Tc in the range of the most recent predetermined period k [days]. −1, Tc) is calculated for each type of X percentile value. Note that the date of the average value A _X (dc-1, Tc) is [dc-1], which represents one day before the date dc of the current day of the calculation time tc to be processed.

Further, the integrated value index calculation unit 11e allows the standard deviation σ _X (dc) of the integrated value M _X (d, Tc) of the X-percentile value of the current change absolute value in the time period Tc in the range of the most recent predetermined period k [days]. −1, Tc) is calculated for each type of X percentile value. The date of the standard deviation σ _X (dc-1, Tc) is [dc-1], which represents one day before the date dc on the current day of the calculation time tc to be processed.

Specifically, for example, 10 percentile value of the current change absolute value in the process of S3 L _10, 25 three percentile percentile values L _25, 50 percentile value L ₅₀ is determined, 10 percentile value in the process of S5, When the integrated values M ₁₀ , M ₂₅ , and M ₅₀ of the 25th percentile value and the 50th percentile value are calculated, the integrated value of the percentile value is calculated every 10th percentile value, 25th percentile value, and 50th percentile value. Average values A ₁₀ , A ₂₅ and A ₅₀ are calculated, and standard deviations σ ₁₀ , σ ₂₅ and σ ₅₀ are calculated.

Here, in order to calculate the average value A _X (dc−1, Tc) and standard deviation σ _X (dc−1, Tc) of the integrated value M _X (d, Tc) of the X percentile value of the current change absolute value. The value of k [days], which is the period as the range of the integrated value M _X (d, Tc) of the X percentile value of the current change absolute value used, is not limited to a specific value and can be adjusted as appropriate. . Specifically, for example, k can be appropriately adjusted and set in the range of about 5 to 30 [day] (however, it is not limited to 5 to 30 [day]). According to the study, about 14 days is preferable.

Then, the integrated value index calculating unit 11e calculates the average value A _X (dc-1, Tc) and standard deviation σ _X (dc-1, Tc) of the integrated values of the X percentile values of the current change absolute value in the time zone Tc. The value is associated with the household identification information and stored in the memory 15 as combination data with the date d and the time zone T.

  Next, the living part of the resident of the household 9 is estimated by the determination part 11f of the control part 11 (S7).

Specifically, the determination unit 11f reads the accumulated value M _X (dc, Tc) of the X percentile value of the current change absolute value in the time zone Tc stored in the memory 15 in the process of S5 from the memory 15, and The average value A _X (dc-1, Tc) and standard deviation σ _X (dc-1, Tc) of the integrated values of the X percentiles of the current change absolute value in the time zone Tc stored in the memory 15 in the process of S6 A value is read from the memory 15.

Then, the determination unit 11f generates the integrated value M _X (dc, Tc) of the X percentile value of the absolute value of the current change in the time zone Tc and the average value A _X (dc-1, Tc) of the integrated value of the X percentile value. Contrast.

Specifically, for each type of the X percentile value, it is determined whether or not the integrated value M _X (dc, Tc) is greatly deviated from the average value A _X (dc-1, Tc) of the integrated value. It is determined which of the relationships 3a to 3c is established.
(Expression 3a) M _X (dc, Tc) <A _X (dc-1, Tc) -pσ _X (dc-1, Tc)
(Equation 3b) A _X (dc-1, Tc) −pσ _X (dc−1, Tc) ≦ M _X (dc, Tc) ≦ A _X (dc−1, Tc) + pσ _X (dc−1, Tc)
(Equation 3c) M _X (dc, Tc)> A _X (dc-1, Tc) + pσ _X (dc-1, Tc)

If the relationship of Equation 3a or Equation 3c holds, the integrated value M _X (dc, Tc) of the X-percentile value of the current change absolute value in the previous time zone Tc (that is, date dc) is the latest. It means that the current change absolute value in the past predetermined period k is greatly deviated from the average value A _X (dc−1, Tc) of the integrated value of the X percentile value. Since it is considered that there is a sudden change (deviation) from the average life pattern of the past past, it is determined that there is a possibility that some abnormal situation / emergency has occurred in the residents of the household 9.

On the other hand, when the relationship of Formula 3b holds, the integrated value M _X (dc, Tc) of the X percentile value of the current change absolute value in the previous time zone (ie, date dc) Tc is the most recent predetermined period. It means that the current change absolute value at k is not greatly deviated from the average value A _X (dc-1, Tc) of the integrated value of the X percentile value. It is considered that there is no change (deviation) from the average life pattern of the household, so there is no possibility that any abnormal situation / emergency has occurred in the residents of the household 9 (ie, healthy and normal) To be judged.

Here, the integrated value standard deviation σ _X (used to determine whether or not the integrated value M _X (dc, Tc) is greatly deviated from the average integrated value A _X (dc-1, Tc). The value of the coefficient p of dc-1, Tc) is not limited to a specific value, and is a value that can capture a sudden change in the usage pattern of the electrical equipment accompanying the indoor activity by a resident of the household 9. It is selected appropriately. Specifically, for example, p can be appropriately selected and set in the range of about 1 to 5 (however, it is not limited to 1 to 5). In this case, a good judgment result can be obtained.

  In addition, it is determined whether the relationship among the mathematical formulas 3a to 3c is established for each type of the X percentile value, and whether there is a possibility that some abnormal situation or emergency situation has occurred in the resident of the household 9 When the judgment is made, if the relationship of the mathematical expression 3a or the mathematical expression 3c is established with respect to the X percentile value index of a part (including only one) of the plurality of X percentile value types, the household It is possible to determine that there is a possibility that some kind of abnormal situation / emergency has occurred in the 9 residents, and for the indicators of all X percentile values of a plurality of types of X percentile values, Only when the relationship of Formula 3c is established, it is judged that there is a possibility that some abnormal situation / emergency has occurred in the residents of household 9 It may be.

  Then, the control unit 11 inputs the measurement data of the total load current from the measurement unit 1 that is a trigger for starting the processing from S2 to the measurement value data server 2 (in other words, records in the total load current database 18). I will wait for you.

  Since it can be determined whether or not an abnormal situation has occurred in the resident of the household 9 by the above processing, the household 9 is composed of, for example, a single household of the elderly or a disabled person, an elderly person, and a disabled person. The present invention can be applied as a mechanism for confirming the safety of the resident of the household 9 by extracting the household and allowing the estimation device 10 of the living situation to further include the notification unit 11g in the control unit 11.

  Specifically, when the relationship of Formula 3a or Formula 3c is established, the notification unit 11g transmits warning information to the external device 5 used by the family of the resident of the household 9, a doctor in charge, or the like. The family, the doctor in charge, or the like who has received the warning information calls the resident to make a safety check or visits the household 9 to check the safety. As a result, when an abnormal situation occurs in an elderly single-person household or an elderly-only household, it becomes possible to prevent the situation from worsening without anyone being aware of it. Only the elderly and the elderly can ensure the safety of their households.

  As can be seen from the above, the final time of each time zone when the time zone of the day is divided is the time when it is determined whether or not an abnormal situation has occurred in the household 9.

According to the living state estimating method, estimating apparatus, and estimating program of the present invention configured as described above, the integration of the X percentile value of the absolute value of current change used when estimating the living state of the resident of the household 9 Since the average value A _X (dc−1, Tc) of the values is set for each household 9 by the total load current I (d, t) measured in the household 9, The index value for estimating the living situation can be automatically set to reflect the characteristics of the use of electrical equipment for each household 9 (in other words, the life pattern). It is not necessary to perform work that requires labor and experience of setting values, and for example, even when many households are subject to estimation simultaneously, it is possible to quickly set appropriate index values for individual households. It becomes possible.

  According to the estimation method, estimation apparatus, and estimation program of the living situation of the present invention, the usage pattern of the electrical equipment in the household 9 changes according to the seasonal change, etc., and the index value can be automatically set. Therefore, it is possible to constantly update the index value reflecting the change in the usage pattern of the electric equipment, and always appropriately estimate the living situation of each household 9. In addition, since the most recent measurement data is used when setting the index value, the index value can be made to reflect the most recent electrical equipment usage pattern of the household 9, so that it is appropriate for the current situation of the household 9 It is possible to always perform a simple estimation.

  According to the living situation estimation method, estimation apparatus, and estimation program of the present invention, the living situation of the resident can be confirmed at a place away from the household 9 where the measuring means 1 is installed. You can check the living situation of whether or not the resident is living steadily without going to 9 directly. For example, it is possible to improve the living safety of single elderly households or households only for the elderly. become. Moreover, since the situation of a plurality of households can be collected and checked in a single living situation estimation device 10, whether the residents of the plurality of households live in a healthy manner with little effort in normal operation. The living situation estimation system can be confirmed, and the disability of the living situation estimation system is reduced to promote the spread of the living situation estimation system. It is possible to improve the performance.

  Although the above embodiment is an example of a preferred embodiment of the present invention, the embodiment of the present invention is not limited to the above embodiment, and various modifications can be made without departing from the gist of the present invention. It is. For example, in the above-described embodiment, the measurement value data server 2 and the life situation estimation device 10 are provided as independent devices, and the life situation estimation device 10 is transmitted from the measurement means 1 via the measurement value data server 2. However, the present invention is not limited to this, and the measurement data is directly input from the measurement means 1 to the living state estimation device 10 (in other words, the measurement value data server 2 and The living situation estimation apparatus 10 is configured as an integrated apparatus. Specifically, the storage unit 12 of the living situation estimation apparatus 10 is used as a storage apparatus that accumulates measurement data transmitted from the measuring means 1). You may do it. Furthermore, the measurement data transmitted from the measuring means 1 may be routed through any route / apparatus as long as it is input to the living situation estimation apparatus 10.

  In the above-described embodiment, the current change absolute value data is recorded (saved) in the calculated value data file 19 stored (saved) in the storage unit 12 or the calculated value data server 2 or other appropriate storage device / storage medium. However, the present invention is not limited to this, and it may be recorded (saved) in the memory 15. In the above-described embodiment, various calculation results are stored in the memory 15 and appropriately read. However, the present invention is not limited to this, and the storage unit 12 or the calculation value data server 2 or other appropriate storage device / storage It may be recorded (saved) in a data file stored (saved) on a medium and read as appropriate.

In the above-described embodiment, a plurality of (three specific examples) X percentile values L _X of current change absolute values are extracted in the process of S3, and the processes of S4 and S5 are performed for each type of X percentile value. Although described above, the present invention is not limited to this, and one or two X percentile values L _X of the current change absolute value are extracted in the process of S3, and the processes of S4 and S5 are performed on one or two X percentile values. Alternatively, four or more X percentile values L _X of the current change absolute value may be extracted in the process of S3, and the processes of S4 and S5 may be performed on the four or more X percentile values. Furthermore, in the above-described embodiment, 10, 25, and 50 are given as examples of X [%] values defining the percentile value extracted as the X percentile value L _X of the current change absolute value. The present invention is not limited to these, and any number of X values may be used, whether it is the 5th percentile value or the 30th percentile value.

  Further, in the above-described embodiment, the time zone of the day is divided in advance, and it is determined whether there is a possibility that any abnormal situation / emergency situation has occurred in the resident of the household 9 for each time zone. However, the present invention is not limited to this, and the time zone of the day may not be divided in advance. The processing contents in this case will be described below with reference to FIG.

First, the processing from S1 to S3 is performed in the same manner as in the above-described embodiment. However, the X percentile value L _X of the current change absolute value in the range of the most recent predetermined time m [minutes] determined in the process of S3 is the time period T of the day divided in advance as in the above embodiment. Without the attribute, it is simply stored as L _X (dc, tc) in the memory 15 as combination data of date d and time t in association with the identification information of the household.

And the process corresponding to the process of S5 in the above-mentioned embodiment is performed as a process of S4 ', without performing the process of S4 in the above-mentioned embodiment. At this time, the integrated value M _X (dc, tc) of the X percentile value L _X (dc, t) of the current change absolute value in the latest past predetermined range S [time] is calculated by the percentile value integrating unit 11 d of the control unit 11. (S4 ').

Specifically, the percentile value integrating unit 11d uses the X-percentile value data of the current change absolute value stored in the memory 15 in the process of S3 (specifically, the X-percentile value L _X (d , T) and the date d / time t), the X percentile value L _X (dc, t) of the predetermined past range S [time] from the calculation time tc is read.

Then, the percentile value accumulating unit 11d accumulates the X percentile value L _X (dc, t) of the current change absolute value in the latest past predetermined range S [time] for each type of the X percentile value. Note that the time t of the X percentile value L _X (dc, t) of the current change absolute value is made the same as the calculation time tc determined in the process of S2-2.

Here, the time as the range of the X percentile value L _X (d, t) of the absolute value of current change used to calculate the integrated value M _X (dc, tc) of the X percentile value of the current change absolute value. The value of S [time] is not limited to a specific value, and can be adjusted as appropriate. Specifically, for example, S can be appropriately adjusted and set in the range of about 3 to 9 [hours] (however, it is not limited to 3 to 9 [hours]). According to examination, about 6 [hour] is preferable. The time S [hour] related to the calculation of the integrated value M _X of the X percentile value of the current change absolute value is set to a value larger than the time m [minute] related to the determination of the X percentile value L _X of the current change absolute value. (Ie, S [hour]> m [minute]).

Further, the integrated value M _X of the X percentile value of the current change absolute value in the most recent past predetermined range S [time] is used as an index representing the amount of operation of the electric device by the resident of the household 9 in the most recent past predetermined time m minutes. This is the total value of the X percentile values L _X of the current change absolute values extracted in the predetermined range S [time]. Therefore, the amount of activity on the electrical equipment by the resident of the household 9 in the most recent predetermined range S [time] is large. (Total amount, corresponding to the integral value).

Then, the percentile value integrating unit 11d associates the integrated value M _X (dc, tc) of the X-percentile value of the current change absolute value in the most recent predetermined range S [time] with the identification information of the household as the date d · It is stored in the memory 15 as combination data with the time t.

Next, a process corresponding to the process of S6 in the above-described embodiment is performed as the process of S5 ′. In other words, the integrated value index calculating unit 11e of the control unit 11 uses the X-percentile integrated value M _X (dc, tc) of the current change absolute value based on the latest past predetermined range k [days]. An average value A _X and standard deviation σ _X of the integrated values of the X percentile values of the current change absolute value of S [time] are calculated (S5 ′).

Specifically, the integrated value index calculation unit 11e causes the most recent k days (that is, dates [dc-1], [dc-2], [dc-3]) from the current day (that is, date dc) at the calculation time tc. ,..., [Dc−k]), the integrated value M _X (d, tc) of the X-percentile value of the current change absolute value in the most recent predetermined range S [time] is S4 ′ up to the previous day. In the processing, data of the integrated value of the X percentile value of the current change absolute value stored in the memory 15 (specifically, the value of the integrated value M _X (d, t) of the X percentile value and the date d and time t) (Combination data). In other words, the integrated value M _X (d, tc) in the range of the latest past k days is read in which the time t is the calculated time tc.

Then, the integrated value index calculation unit 11e calculates the average of the integrated values M _X (d, tc) of the X percentile values of the current change absolute values in the most recent past predetermined range S [time] in the range of the most recent past predetermined period k [days]. A value A _X (dc-1, tc) is calculated for each type of X percentile value.

Further, the integrated value index calculation unit 11e uses the standard value of the integrated value M _X (d, Tc) of the X-percentile value of the current change absolute value in the latest past predetermined range S [time] in the range of the latest past predetermined period k [day]. Deviation σ _X (dc-1, tc) is calculated for each type of X percentile value.

The method and concept of setting the date of the average value A _X (dc-1, tc), the date of the standard deviation σ _X (dc-1, tc), and the value of k [day] are described in the above embodiment. It is the same.

The integrated value index calculation unit 11e then calculates the average value A _X (dc-1, tc) and standard deviation σ _X (dc−) of the X-percentile value of the absolute value of current change in the most recent predetermined range S [time]. 1, tc) is associated with household identification information and stored in the memory 15 as combination data of the date d and the time zone T.

  Next, a process corresponding to the process of S7 in the above-described embodiment is performed as the process of S6 ′. That is, the living state of the resident of the household 9 is estimated by the determination unit 11f of the control unit 11 (S6 ′).

Specifically, the integrated value M _X (dc, tc) of the X percentile value of the current change absolute value in the latest past predetermined range S [time] stored in the memory 15 in the process of S4 ′ is stored in the memory by the determination unit 11f. 15, and the average value A _X (dc−1, tc) of the integrated values of the X percentile values of the current change absolute value in the latest past predetermined range S [time] stored in the memory 15 in the process of S 5 ′ and The value of the standard deviation σ _X (dc−1, tc) is read from the memory 15.

Then, the determination unit 11f causes the integrated value M _X (dc, tc) of the X percentile value of the absolute value of the current change in the latest past predetermined range S [time] and the average value A _X (dc-1) of the integrated value of the X percentile value. , Tc).

Some abnormal situation / emergency occurs in the resident of the household 9 based on the concept of comparison between the integrated value M _X (dc, tc) and the average value A _X (dc-1, tc) of the integrated value and the comparison result The concept of determining whether or not there is a possibility is that the integrated value M _X (dc, Tc), average value A _X (dc-1, Tc), standard deviation σ _X (dc) of the time zone Tc in the above-described embodiment. −1, Tc) is the integrated value M _X (dc, tc), average value A _X (dc−1, tc), standard deviation σ _X (dc−1, tc) of the most recent predetermined range S [time]. For example, the description is the same as in the above-described embodiment.

  Then, the control unit 11 inputs the measurement data of the total load current from the measurement unit 1 that is a trigger for starting the processing from S2 to the measurement value data server 2 (in other words, records in the total load current database 18). I will wait for you.

  As described above, when the processing from S4 ′ to S6 ′ is performed following the processing from S1 to S3 without dividing the time zone of the day in advance, the time zone Tc in the above-described embodiment is set to the latest past. It is always moved as a predetermined range S [time]. Therefore, the timing at which it is determined whether or not an abnormal situation has occurred in the household 9 is only the last time of each time zone in which the day is divided in the above-described embodiment, whereas the time zone of the day is When it is not classified in advance, it is every calculation time tc. Therefore, the estimation of the living situation of the resident of the household 9 will be made in detail in time. For example, when an abnormal situation occurs in an elderly single-person household or an elderly-only household, this will take a long time. It becomes possible to prevent the situation from becoming worse without being known by anyone, and the living safety of the elderly single-person household or the elderly-only household, etc. is further ensured.

  An embodiment performed for verifying the determination accuracy of the method for estimating the living situation of the present invention will be described with reference to FIGS.

  In the present example, total load current data for one minute for 60 days obtained from a single elderly household was used.

  It has been found in advance that a single person living alone in the above-mentioned household is admitted on the 60th day from the 57th day of the 60th day.

  Therefore, from the viewpoint of ensuring reliability when the present invention is applied, for example, as a mechanism for confirming the safety of a household resident, it is possible to reliably detect abnormal situations / emergencies due to poor physical condition after the 57th day in the above household. is important.

In this example, the time zone of the day was divided into four sections every 6 hours as follows.
First time zone T ₁ : 3:00 am to 9:00 am Second time zone T ₂ : 9:00 am to 3:00 pm Third time zone T ₃ : 3:00 pm to 9:00 pm Fourth time zone T ₄ : Afternoon 9am-3am the next day

In this embodiment, the 10th percentile value L ₁₀ (d, T, t) of the current change absolute value in the range of the latest predetermined time m = 30 [minutes] is determined based on the current change absolute value ΔI (d, t). (S3) and the integrated value of the 10th percentile value L ₁₀ (d, T, t) of the current change absolute value in each time zone T (specifically, T ₁ , T ₂ , T ₃ , T ₄ ) M ₁₀ (d, T) was calculated (S5).

Next, based on the integrated value M ₁₀ (d, T) of the 10th percentile value of the current change absolute value, the average of the integrated value of the 10th percentile value in each time zone T in the range of the most recent predetermined period k = 14 [days]. The value A ₁₀ (d-1, T) and the standard deviation σ ₁₀ (d-1, T) were calculated (S6).

  Then, using Equations 3a to 3c, the living situation of a single elderly person who is a resident of the household was estimated (S7). In this embodiment, the coefficient p = 3 in the mathematical expressions 3a to 3c is set.

The integrated value M ₁₀ (d, T) of the 10th percentile value of the current change absolute value in each time zone T (T ₁ , T ₂ , T ₃ , T ₄ ) and the threshold values in the equations 3a and 3b. The result shown in FIG. 5 was obtained as a time transition over 60 days with “A ₁₀ (d−1, T) −3σ ₁₀ (d−1, T)”. FIG. 5 also shows a time transition of an alarm indicating whether or not an abnormal situation / emergency may occur. The alarm indicates that “0” is determined that there is no possibility that any abnormal situation / emergency has occurred in the resident of the household (that is, healthy and normal), and “1”. Represents that there is a possibility that some abnormal situation / emergency has occurred in the resident of the household.

FIG. 6 is a diagram showing extracted from the 56th day to the 60th day in FIG. 6, the integrated value M ₁₀ (d, T) in a first time period T ₁ of the 57 day approaches to the threshold [A ₁₀ -3σ _10], the second time zone T ₂ integrated value M ₁₀ ( d, T) was below the threshold value, and it was confirmed that there was a possibility of occurrence of an abnormal situation / emergency situation. After that, the judgment that there was a possibility of occurrence of an abnormal situation / emergency situation was continued for three hours, and after that, it returned to the judgment that there was no possibility of occurrence of an abnormal situation / emergency situation. it was confirmed that the third time period T ₃ of being generated potential there and again judgment of the abnormal situation and emergency situation.

  The above-mentioned household resident alone living on the 60th day calls himself an ambulance and is hospitalized. By applying the present invention as a mechanism for confirming the safety of the household resident, the 57th day or later At the time of the 58th day, it was considered that the third party was able to grasp the occurrence of abnormal situations and emergencies of the elderly living alone, and was able to respond to the situation that occurred to the elderly living alone at an early stage. It was.

  From this result, the living situation estimation method of the present invention can reliably detect the occurrence of an abnormal situation / emergency situation in a resident of a household, and has sufficient judgment accuracy to withstand practical use. confirmed.

DESCRIPTION OF SYMBOLS 1 Measuring means 2 Measured value data server 9 Household 10 Living condition estimation apparatus

Claims (18)

  A step of measuring the total load current of the whole household by the measuring means, a step of calculating an absolute value of the change of the total load current using the measurement data of the total load current, and a range of the most recent predetermined time Extracting an X percentile value (a value of X is preset) in an ascending order or descending order of a plurality of absolute values of the change in the total load current at the current value and setting it as the X percentile value of the current change absolute value; A step of calculating an integrated value of the X percentile value of the current change absolute value in a preset time zone by dividing the time zone of the day, and using the integrated value of the X percentile value of the current change absolute value The average value and the standard deviation of the integrated values of the X percentile values of the current change absolute value in the time period in the most recent predetermined period range are calculated. The integrated value of the X percentile value of the current change absolute value and the average value of the integrated value of the X percentile value of the current change absolute value take into account the standard deviation of the integrated value of the X percentile value of the current change absolute value. And estimating the living situation of the resident of the household.   A step of measuring the total load current of the whole household by the measuring means, a step of calculating an absolute value of the change of the total load current using the measurement data of the total load current, and a range of the most recent predetermined time Extracting an X percentile value (a value of X is preset) in an ascending order or descending order of a plurality of absolute values of the change in the total load current at the current value and setting it as the X percentile value of the current change absolute value; A step of calculating an integrated value of the X percentile value of the current change absolute value in the most recent past predetermined range, and the integrated value of the X percentile value of the current change absolute value using the integrated value of the X percentile value of the current change absolute value A step of calculating an average value and a standard deviation of integrated values of X percentile values of the current change absolute value in a range; and The integrated value of the X percentile value of the absolute value and the average value of the integrated value of the X percentile value of the current change absolute value are compared with each other in consideration of the standard deviation of the integrated value of the X percentile value of the current change absolute value. And a step of estimating a living situation of a resident of the household.   The time zone is divided into four categories from 3 am to 9 am, 9 am to 3 pm, 3 pm to 9 pm, and 9 pm to 3 am on the following day. Item 2. A method for estimating a living situation according to Item 1.   The method for estimating a living situation according to claim 1 or 2, wherein the latest past predetermined time is set in a range of 15 to 60 minutes.   3. The method for estimating a living situation according to claim 1, wherein the predetermined period in the past is set in a range of 5 to 30 days.   The living state estimation method according to claim 2, wherein the most recent past predetermined range is set in a range of 3 to 9 hours.   Means for reading data on the total load current of the entire household measured by the measuring means from a storage device; means for calculating an absolute value of the change in the total load current using the data on the total load current; Means for extracting an X percentile value (X value is preset) in an ascending order or descending order of a plurality of absolute values of a change in the total load current in a time range to obtain an X percentile value of a current change absolute value And means for calculating an integrated value of the X percentile value of the current change absolute value in a preset time zone in which the time zone of the day is divided, and an integrated value of the X percentile value of the current change absolute value Means for calculating an average value and a standard deviation of the integrated values of the X percentile values of the current change absolute value in the time period in the most recent past predetermined period range; Compare the integrated value of the X percentile value of the current change absolute value with the average value of the integrated value of the X percentile value of the current change absolute value, taking into account the standard deviation of the integrated value of the X percentile value of the current change absolute value. And a means for estimating a living situation of a resident of the household.   Means for reading data on the total load current of the entire household measured by the measuring means from a storage device; means for calculating an absolute value of the change in the total load current using the data on the total load current; Means for extracting an X percentile value (X value is preset) in an ascending order or descending order of a plurality of absolute values of a change in the total load current in a time range to obtain an X percentile value of a current change absolute value And means for calculating an integrated value of the X percentile value of the current change absolute value in the most recent past predetermined range, and the integrated past value of the X percentile value of the current change absolute value using the integrated value of the X percentile value of the current change absolute value. Means for calculating an average value and a standard deviation of the integrated values of the X percentile values of the current change absolute value in a predetermined range; Residents in the household comparing the integrated value of the percentile value and the average value of the integrated value of the X percentile value of the current change absolute value with the standard deviation of the integrated value of the X percentile value of the current change absolute value taken into account A living state estimating device characterized by comprising: means for estimating the living state of   The time zone is divided into four categories from 3 am to 9 am, 9 am to 3 pm, 3 pm to 9 pm, and 9 pm to 3 am on the following day. Item 8. The living state estimating apparatus according to Item 7.   9. The living state estimating apparatus according to claim 7, wherein the most recent past predetermined time is set in a range of 15 to 60 minutes.   9. The living state estimation apparatus according to claim 7, wherein the most recent predetermined period is set in a range of 5 to 30 days.   The living state estimation apparatus according to claim 8, wherein the most recent past predetermined range is set in a range of 3 to 9 hours.   Means for reading the data of the total load current of the entire household measured by the measuring means from the storage device, means for calculating the absolute value of the change of the total load current using the data of the total load current, the last past predetermined time Means for extracting an X percentile value (X value is preset) in an ascending order or descending order of a plurality of absolute values of a change in the total load current in a range, and obtaining an X percentile value of a current change absolute value; Means for calculating an integrated value of the X percentile value of the current change absolute value in a time zone set in advance by dividing the time zone of the day; Means for calculating an average value and a standard deviation of integrated values of X percentiles of current change absolute values in the time zone in a period range, the current change absolute The X-th percentile value integrated value of the current change and the average value of the X-percentile value integrated value of the current change absolute value are compared with the standard deviation of the integrated value of the X-percentile value of the current change absolute value. A living situation estimation program for causing a computer to function as a means for estimating the living situation of a resident.   Means for reading the data of the total load current of the entire household measured by the measuring means from the storage device, means for calculating the absolute value of the change of the total load current using the data of the total load current, the last past predetermined time Means for extracting an X percentile value (X value is preset) in an ascending order or descending order of absolute values of a plurality of changes in the total load current in a range, and obtaining an X percentile value of the current change absolute value; Means for calculating an integrated value of the X percentile value of the current change absolute value in the past predetermined range, and current in the latest past predetermined range in the range of the latest past predetermined period using the integrated value of the X percentile value of the current change absolute value Means for calculating an average value and a standard deviation of the integrated values of the X percentile values of the change absolute value, and the X percentile of the current change absolute value The average value of the integrated value of the X-percentile value of the current change absolute value and the average value of the integrated value of the X-percentile value of the current change absolute value are compared with the standard deviation of the integrated value of the X percentile value of the current change absolute value. A living situation estimation program for causing a computer to function as a means for estimating a living situation.   The time zone is divided into four categories from 3 am to 9 am, 9 am to 3 pm, 3 pm to 9 pm, and 9 pm to 3 am on the following day. Item 14. The living situation estimation program according to Item 13.   The life situation estimation program according to claim 13 or 14, wherein the latest past predetermined time is set in a range of 15 to 60 minutes.   The life situation estimation program according to claim 13 or 14, wherein the most recent predetermined period is set in a range of 5 to 30 days.   The life situation estimation program according to claim 14, wherein the most recent past predetermined range is set in a range of 3 to 9 hours.

Images