Analysis device

The analysis device uses power and temperature data to decompose consumption into multiple components, addressing the limitations of existing methods by accurately analyzing power usage in facilities with continuous air conditioning, facilitating energy conservation.

JP7583402B2Active Publication Date: 2024-11-14OSAKA UNIVERSITY +1
View PDF 5 Cites 0 Cited by

Patent Information

Application Number
JP2020132435
Authority / Receiving Office
JP · JP
Patent Type
Patents
Current Assignee / Owner
Filing Date
2020-08-04
Publication Date
2024-11-14
Estimated Expiration
2040-08-04

AI Technical Summary

Technical Problem

Existing methods for analyzing power consumption, such as those described in Non-Patent Document 1, are limited to buildings where air conditioning equipment is not in operation, making them unsuitable for facilities like accommodation and medical facilities where air conditioning is constantly used.

Method used

An analysis device that utilizes power consumption data and outside temperature data to decompose power consumption into multiple components, including base, activity-induced fixed, and seasonal variation components, using regression analysis to calculate and analyze power consumption accurately in facilities with continuous air conditioning.

Benefits of technology

Enables detailed analysis of power consumption in various consumer types, allowing for effective energy conservation strategies by identifying and decomposing power usage into distinct components, including those dependent and independent of human activity and seasonal variations.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 0007583402000001
    Figure 0007583402000001
  • Figure 0007583402000002
    Figure 0007583402000002
  • Figure 0007583402000003
    Figure 0007583402000003
Patent Text Reader

Abstract

To provide technology with which it is possible to analyze the consumed amount of electric power of various users.SOLUTION: An analysis unit generates, using electric power amount data and ambient temperature, first correlation information that indicates a correlation between an ambient temperature and the minimum consumed amount of electric power at the ambient temperature. The analysis unit decomposes, using the first correlation information and ambient temperature data, the consumed amount of electric power in a target time slot of each day that is included in an analysis target period into a base component 21 (first component) independent of human activities and a second component 22. The analysis unit generates, using ambient temperature data, second correlation information that indicates a correlation between the ambient temperature and the second component 22 in the target time slot, with regard to a plurality of days included in the analysis target period. The analysis unit decomposes, using the second correlation information and ambient temperature data, the second component 22 in the target time slot of each of the plurality of days into an activity-induced fixed component 23 (third component) that does not change with the ambient temperature and a fourth component 24.SELECTED DRAWING: Figure 5
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] The present disclosure relates to a technique for analyzing power consumption of a consumer. [Background technology]

[0002] Conventionally, technologies have been developed to analyze the amount of power consumption by use based on the amount of power consumption by consumers. For example, Non-Patent Document 1 discloses a method for decomposing the power demand of an office building into a base component that is not dependent on human activity, an air conditioning component associated with the operation of air conditioning equipment, and an activity component caused by human activity other than air conditioning. The analysis results are used for energy conservation. [Prior art documents] [Non-patent literature]

[0003] [Non-Patent Document 1] Hiroki Ohshima et al., "Development of a method for evaluating the actual energy usage in buildings using total power consumption by time of day," Journal of the Society of Heating, Air-Conditioning and Sanitary Engineers of Japan, Vol. 264, March 2019, pp. 13-21 Summary of the Invention [Problem to be solved by the invention]

[0004] The technology described in Non-Patent Document 1 analyzes the energy consumption by purpose of a consumer by using the energy consumption on days when the air conditioning equipment is not operated. In other words, the technology described in Non-Patent Document 1 targets buildings where there are days when the air conditioning equipment is not operated. Therefore, the technology described in Non-Patent Document 1 cannot be applied to the analysis of the energy consumption of consumers who use buildings where the air conditioning equipment is constantly operating, such as accommodation facilities and medical facilities.

[0005] The present disclosure has been made in consideration of the above-mentioned problems, and has an object to provide a technique capable of analyzing the amount of power consumption of various consumers. [Means for solving the problem]

[0006] According to one aspect, an analysis device includes a first acquisition unit, a second acquisition unit, and an analysis unit. The first acquisition unit acquires power data indicating power consumption of a consumer measured for each time period during an analysis period. The second acquisition unit acquires outdoor temperature data indicating outdoor temperatures for each time period in the consumer's area during the analysis period. The analysis unit analyzes power consumption for each time period on each day included in the analysis period. The analysis unit uses the power data and the outdoor temperature data to generate first correlation information indicating a correlation between the outdoor temperature and the minimum power consumption at the outdoor temperature. The analysis unit uses the first correlation information and the outdoor temperature data to decompose the power consumption for the target time period on each day included in the analysis period into a first component that does not depend on human activity and a second component other than the first component. The analysis unit uses the outdoor temperature data to generate second correlation information indicating a correlation between the outdoor temperature for the target time period and the second component for a plurality of days included in the analysis period. The analysis unit uses the second correlation information and the outside temperature data to decompose the second component in the target time period for each of the multiple days into a third component that does not vary according to the outside temperature, and a fourth component other than the third component.

[0007] In the above analysis device, the second correlation information indicates a first formula obtained by regression analysis with the outside temperature as the first explanatory variable and the second component as the first objective variable. The analysis unit calculates the outside temperature influence component by subtracting the minimum value of the first objective variable when the first explanatory variable is changed in the first formula from the value of the first objective variable obtained by substituting the outside temperature during the target time period of each of the multiple days for the first explanatory variable in the first formula. The analysis unit calculates the difference between the second component and the outside temperature influence component for the target time period of each of the multiple days, and calculates the third component based on the difference.

[0008] In the above analysis device, the analysis unit classifies each of the multiple days into one of multiple groups defined by at least one of the day of the week, whether it is a holiday, and whether it is a public holiday. The analysis unit determines a representative value of the differences between the days belonging to each of the multiple groups as the third component in the target time zone of the day belonging to that group.

[0009] In the above analysis device, the second correlation information indicates a first equation obtained by regression analysis with the outside air temperature as the first explanatory variable and the second component as the first objective variable. The analysis unit determines, as the third component, the minimum value of the first objective variable when the first explanatory variable is changed in the first equation.

[0010] In the above analysis device, the analysis unit divides the analysis period into a first period and a second period in which the outside temperature is lower than that of the first period. A plurality of days belong to one of the first period and the second period.

[0011] In the above analysis device, the analysis unit uses the outside temperature data to generate third correlation information indicating a correlation between the outside temperature in the target time zone and the fourth component for multiple days. The analysis unit uses the third correlation information to decompose the fourth component in the target time zone for each of the multiple days into a fifth component and a sixth component other than the fifth component. The third correlation information indicates a third equation obtained by a regression analysis with the outside temperature as the second explanatory variable and the fourth component as the second objective variable. The analysis unit calculates the fifth component by subtracting the minimum value of the second objective variable when the second explanatory variable is changed in the third equation from the value of the second objective variable obtained by substituting the outside temperature in the target time zone for each of the multiple days for the second explanatory variable in the third equation.

[0012] The above-mentioned analysis device further includes a notification unit that notifies of an abnormality based on the sixth component. In the above analysis device, the notification unit predicts a predicted value of the sixth component in a detection target period, which is composed of one or a plurality of consecutive time periods, using a model that predicts the predicted value of the sixth component in the detection target period from the analysis result of the sixth component in a time period prior to the detection target period. The notification unit detects the occurrence of an abnormality when a difference between a representative value of the analysis result of the sixth component in the detection target period and the predicted value is greater than a threshold. The threshold is determined according to a distribution of differences between a representative value of the analysis result of the sixth component for each unit time having the same time length as the detection target period, which is included in the analysis target period, and a predicted value of the sixth component in the unit time predicted using the model.

[0013] In the above analysis device, the notification unit detects the occurrence of an abnormality when a representative value of the analysis result of the sixth component in a detection period consisting of one or multiple consecutive time periods is greater than a threshold value. The threshold value is determined according to a distribution of the representative value of the analysis result of the sixth component for each unit time of the same length as the detection period, which is included in the analysis period.

[0014] In the above-described analysis device, the notification unit detects the occurrence of an abnormality when the minimum value of the analysis result of the sixth component in all time periods on the target day is greater than a threshold determined from the first component.

[0015] In the above-described analysis device, the notification unit detects the occurrence of an abnormality when the integrated value or average value of the analysis result of the sixth component for all time periods on a target day is greater than a threshold determined from the third component.

[0016] In the above-described analysis device, the notification unit detects the occurrence of an abnormality when the analysis result of the sixth component in the target time period on the target day is greater than a threshold determined from the third component.

[0017] In the above-mentioned analytical device, the notification unit detects the occurrence of an abnormality when the degree of change in the analysis result of the sixth component in response to a change in outside air temperature during one or more predetermined time periods on a day belonging to a period of at least two days in length is greater than a threshold value determined from the fifth component.

[0018] In the above analysis device, the first correlation information indicates a fifth equation obtained by regression analysis with the outside temperature as a third explanatory variable and the minimum power consumption as a third objective variable. The analysis unit determines the value of the third objective variable obtained by substituting the outside temperature in the target time period of each of the multiple days for the third explanatory variable in the fifth equation as the first component in the target time period of each of the multiple days.

[0019] In the above analysis device, the analysis unit determines, as the seventh component, the minimum value of the third objective variable when the third explanatory variable is changed in the fifth formula. The analysis unit decomposes the first component in the target time slot of each day included in the analysis period into the seventh component and an eighth component other than the seventh component.

[0020] According to another aspect, the analysis method includes the steps of acquiring power data indicating power consumption of a consumer measured for each time period during an analysis period, acquiring outdoor temperature data indicating outdoor temperatures for each time period in the consumer's area during the analysis period, and analyzing power consumption for each time period on each day included in the analysis period. The analyzing step includes a step of generating first correlation information indicating a correlation between the outdoor temperature and the minimum power consumption at the outdoor temperature using the power data and the outdoor temperature data. The analyzing step further includes a step of decomposing the power consumption in the target time period on each day included in the analysis period into a first component that does not depend on human activity and a second component that depends on human activity using the first correlation information and the outdoor temperature data. The analyzing step further includes a step of generating second correlation information indicating a correlation between the outdoor temperature in the target time period and the second component for a plurality of days included in the analysis period using the outdoor temperature data. The analyzing step further includes a step of decomposing the second component into a third component that does not vary depending on the outdoor temperature and a fourth component other than the third component using the second correlation information and the outdoor temperature data.

[0021] In another aspect, a program causes a computer to execute each step of the above analysis method. Effect of the Invention

[0022] According to the technology of the present disclosure, the power consumption of various consumers can be analyzed. [Brief description of the drawings]

[0023] [Figure 1] FIG. 1 is a diagram showing an outline of a system including an analysis device according to an embodiment of the present invention. [Diagram 2]FIG. 2 is a diagram illustrating a hardware configuration of an analysis apparatus. [Diagram 3] FIG. 2 is a diagram illustrating a functional configuration of an analysis device. [Figure 4] 2 is a diagram showing classifications of a plurality of electrical devices 220 installed in a building 200. FIG. [Diagram 5] FIG. 2 is a diagram showing a plurality of components constituting the amount of power consumption. [Figure 6] 4 is a flowchart showing an example of a process flow in the analysis device 100. [Figure 7] FIG. 2 is a diagram for explaining a method for calculating a base component 21. [Figure 8] FIG. 2 is a diagram for explaining a method for calculating a base fixed component 27 and a base seasonal fluctuation component 28. [Figure 9] FIG. 11 is a diagram showing a method for calculating the outside air temperature influence component included in the second component 22. [Figure 10] FIG. 13 is a diagram showing the distribution of the difference between the second component 22 and the outside air temperature influence component. [Figure 11] FIG. 13 is a diagram showing an example of a change in the activity-based fixed component 23 over a one week period. [Figure 12] 13 is a diagram showing the distribution of a fourth component 24 obtained by subtracting an activity-attributable fixed component 23 from a second component 22, with respect to outside air temperature. [Figure 13] FIG. 13 is a diagram showing an example of a regression curve 34 indicating the correlation between the outside air temperature and a fourth component. [Figure 14] FIG. 13 is a diagram showing the fluctuation of a residue component 26 during a part of an analysis period. [Figure 15] FIG. 13 is a diagram showing an example of a display of analysis results. [Figure 16] FIG. 11 is a diagram showing another example of display of the analysis results. [Figure 17] FIG. 13 is a diagram showing an example of changes in activity-induced fixed components 23 for multiple offices. [Figure 18] FIG. 13 is a diagram showing an example of changes in activity-induced fixed components 23 for multiple general merchandise stores. [Figure 19] FIG. 13 is a diagram showing an example of changes in activity-induced fixed components 23 for a plurality of accommodation facilities. [Figure 20]A diagram for explaining an autoregressive model. [Figure 21] A diagram showing an example of the probability distribution of the difference between the representative value and the predicted value of the analysis result of the residue component 26 for each unit time. [Figure 22] A diagram showing the component analysis result of the power consumption and the predicted value of the residue component 26. [Figure 23] A diagram showing an example of the probability distribution of the representative value of the residue component 26 per unit time. [Figure 24] A diagram showing the analysis result of the residue component 26 for each unit time. [Diagram 25] A diagram showing the phenomenon that causes the increase of the residue component 26 and the determination method of the presence or absence of the phenomenon. [Figure 26] A diagram showing the integrated value of the analysis result of the residue component 26. [Figure 27] A diagram showing the component analysis result of the power consumption. [Figure 28] A diagram showing the component analysis result of the power consumption and the threshold Thc. [Figure 29] A diagram showing the temperature sensitivity of the residue component 26.

Embodiments for Carrying Out the Invention

[0024] Hereinafter, embodiments according to the present invention will be described with reference to the drawings. In the following description, the same parts and components are denoted by the same reference numerals, and overlapping descriptions will not be repeated. Note that the embodiments and each modification described below may be selectively combined as appropriate.

[0025] <A. System Configuration> With reference to FIG. 1, the configuration of a system including an analyzer according to the present embodiment will be described. FIG. 1 is a diagram showing an overview of a system including an analyzer according to the present embodiment. As shown in FIG. 1, the system SYS includes an analyzer 100, a building 200 to be analyzed, and a server device 300.

[0026] The building 200 is used by consumers and includes, for example, stores, offices, medical facilities, accommodation facilities, etc. Note that the building 200 may refer to the entire building or a part of the building (for example, one floor). A smart meter 210 is installed in the building 200. The smart meter 210 measures and records the power consumption of the consumer for each time period. Specifically, for each time period obtained by dividing the time from 0:00 to 24:00 into a predetermined time length (for example, 30 minutes or 1 hour), the total power consumption consumed by various electrical devices 220 installed in the building 200 is measured.

[0027] The smart meter 210 communicates with the server device 300 and transmits power consumption data indicating the measured power consumption to the server device 300. The server device 300 stores the power consumption data received from the smart meter 210.

[0028] The analysis device 100 is connected to the network and acquires the power consumption data of the building 200 from the server device 300 via the network. The analysis device 100 analyzes the power consumption state in the building 200 based on the power data. The analysis device 100 may generate information regarding energy conservation using the analysis result. The generated information is notified to the consumer. By checking the notified information, the consumer can perform actions suitable for energy conservation.

[0029] <B. Hardware Configuration of Analysis Device> FIG. 2 is a diagram showing the hardware configuration of the analyzer. As shown in FIG. 2, the analyzer 100 mainly includes a processor 101 that executes a program 110, a ROM (Read Only Memory) 102 that stores data non-volatilely, a RAM (Random Access Memory) 103 that stores data generated by the execution of the program 110 by the processor 101 or data input via an input device volatilely, a hard disk (HDD) 104 that stores data non-volatilely, a communication IF (Interface) 105, an operation key 106, a power supply circuit 107, and a display 108. Each component is connected to each other by a data bus. Note that the communication IF 105 is an interface for performing communication with other devices.

[0030] The processing in the analyzer 100 is realized by each hardware and the program 110 executed by the processor 101. Such a program 110 is stored in the HDD 104 in advance. However, the program 110 may be stored in other storage media and distributed as a program product. Alternatively, the program 110 may be provided as a program product that can be downloaded by an information provider connected to the so-called Internet. Such a program 110 is read from the storage media by a reading device or downloaded via the communication IF 105 or the like and then stored in the HDD 104. The processor 101 reads the program 110 from the HDD 104 and executes the program 110.

[0031] <C. Functional Configuration of Analyzer> FIG. 3 is a diagram showing the functional configuration of the analyzer. As shown in FIG. 3, the analyzer 100 includes a storage unit 10, a first acquisition unit 11, a second acquisition unit 13, an analysis unit 15, and a notification unit 16. The first acquisition unit 11, the second acquisition unit 13, the analysis unit 15, and the notification unit 16 are realized by the processor 101 executing the program 110. The storage unit 10 is realized by the ROM 102 and the RAM 103.

[0032] The first acquisition unit 11 acquires, from the server device 300 via the network, the power consumption data 12 of the consumers during the analysis period for the building 200 (see FIG. 1) to be analyzed. The analysis period is, for example, one year. The first acquisition unit 11 stores the acquired power consumption data 12 in the storage unit 10.

[0033] The second acquisition unit 13 acquires, via the network, the outside air temperature data 14 indicating the outside air temperature for each time zone in the area of the consumers (that is, the area including the building 200 (see FIG. 1)) during the analysis period. The outside air temperature data 14 is distributed from a predetermined institution (for example, the Japan Meteorological Agency). The second acquisition unit 13 stores the acquired outside air temperature data 14 in the storage unit 10.

[0034] The analysis unit 15 analyzes the power consumption for each time zone of each day included in the analysis period using the power consumption data 12 and the outside air temperature data 14. Specifically, the analysis unit 15 performs component analysis of the power consumption.

[0035] The notification unit 16 notifies the analysis result by the analysis unit 15, generates information regarding energy conservation obtained from the analysis result, and notifies the generated information. The notification unit 16 causes, for example, the analysis result and the information regarding energy conservation to be displayed on the display 108 (see FIG. 2).

[0036] <D. Component Analysis of Power Consumption> The plurality of electrical devices 220 (see FIG. 1) installed in the building 200 are classified into season-dependent devices whose power consumption varies according to the outside air temperature and non-season-dependent devices whose power consumption does not vary according to the outside air temperature. Further, the plurality of electrical devices 220 are classified into devices that are always in operation and devices that are in operation when people are active.

[0037] FIG. 4 is a diagram showing classification of a plurality of electrical devices 220 installed in a building 200. As shown in FIG. 4, examples of non-seasonal electrical devices that are constantly operated include a server and a common room ventilation system. Examples of non-seasonal electrical devices that are operated during human activity include lighting fixtures and devices connected to an outlet. Examples of seasonal electrical devices that are constantly operated include an air conditioner, a refrigerator, and a warm water washing toilet seat that are operated 24 hours a day. Examples of seasonal electrical devices that are operated during human activity include an air conditioner and an electric water heater that are used only during limited times.

[0038] The analysis unit 15 decomposes the power consumption of the consumer into a plurality of components, taking into consideration the classification shown in FIG.

[0039] 5 is a diagram showing multiple components that make up the amount of power consumption. The analysis unit 15 decomposes the amount of power consumption 20 for each time period on each day of the analysis period into a base component 21 (also referred to as the "first component") that does not depend on human activity, and a second component 22 other than the base component 21. The base component 21 corresponds to the amount of power consumed by electrical appliances that are constantly in operation.

[0040] The analysis unit 15 decomposes the second component 22 into an activity-induced fixed component 23 (also referred to as the "third component") that does not vary with the outside temperature, and a fourth component 24 that is other than the activity-induced fixed component 23. The activity-induced fixed component 23 corresponds to the amount of power consumed by non-seasonally dependent electrical appliances that are operated during human activity.

[0041] Analysis unit 15 decomposes fourth component 24 into activity-related seasonal fluctuation component 25 (also referred to as "fifth component") that fluctuates according to the outside air temperature, and residual component 26 (also referred to as "sixth component") other than activity-related seasonal fluctuation component 25. Activity-related seasonal fluctuation component 25 corresponds to the amount of power consumed by season-dependent electrical appliances that are operated during human activity.

[0042] The analysis unit 15 decomposes the base component 21 into a base fixed component 27 (also referred to as the "seventh component") that does not vary according to the outside air temperature and a base seasonal variation component 28 (also referred to as the "eighth component") other than the base fixed component 27. The base fixed component 27 corresponds to the amount of power consumed by non-season-dependent electrical equipment that is always in operation. The base seasonal variation component 28 corresponds to the amount of power consumed by season-dependent electrical equipment that is always in operation.

[0043] <E. Flow of processing of the analysis device> FIG. 6 is a flowchart showing an example of the flow of processing in the analysis device 100. As shown in FIG. 6, first, the first acquisition unit 11 of the analysis device 100 acquires power consumption data 12 indicating the amount of power consumed for each time period in the analysis target period for the consumer to be analyzed (step S1). The acquired power consumption data 12 is stored in the storage unit 10.

[0044] Next, the second acquisition unit 13 of the analysis device 100 acquires outside air temperature data 14 indicating the outside air temperature for each time period in the area of the consumer in the analysis target period (step S2). The acquired outside air temperature data 14 is stored in the storage unit 10. Note that step S2 may be executed before step S1.

[0045] Next, the analysis unit 15 of the analysis device 100 decomposes the power consumption 20 for each time period on each day included in the analysis target period into a base component 21 (see FIG. 5) that does not depend on human activities and a second component 22 other than the base component 21 (step S3).

[0046] Next, the analysis unit 15 decomposes the second component 22 into an activity-caused fixed component 23 and a fourth component 24 other than the activity-caused fixed component 23 (step S4).

[0047] Next, the analysis unit 15 decomposes the fourth component 24 into an activity-caused seasonal variation component 25 and a residue component 26 other than the activity-caused seasonal variation component 25 (step S5).

[0048] Furthermore, in parallel with steps S4 and S5, the analysis unit 15 decomposes the base component 21 into a base fixed component 27 and a base seasonal variation component 28 other than the base fixed component 27 (step S6).

[0049] After the completion of steps S5 and S6, the notification unit 16 of the analyzer 100 notifies the analysis result (step S7). In step S7, the notification unit 16 may generate information regarding energy saving from the analysis result and notify the generated information. After step S7, the analyzer 100 ends the process.

[0050] <F. Calculation method of base component> With reference to FIG. 7, a specific processing method of step S3 will be described. FIG. 7 is a diagram for explaining the calculation method of the base component 21.

[0051] The analysis unit 15 generates a unit data set in which the power consumption and the outside air temperature in each time zone of each day included in the analysis target period are associated with each other using the power consumption data 12 and the outside air temperature data 14. For example, when the number of days included in the analysis target period is 365 and each day is divided into 48 time zones every 30 minutes, the analysis unit 15 generates 365 × 48 = 17520 unit data sets.

[0052] The base component 21 corresponds to the power consumption by the electrical equipment that is always in operation. Therefore, the calculation accuracy of the base component 21 can be improved by using the power consumption in the time zone when people are not active. Thus, the analysis unit 15 extracts the unit data sets corresponding to the predetermined time zones of each day included in the analysis target period. The predetermined time zone includes the time zone when people are not active (for example, the late night time zone from 0:00 to 3:00), and is, for example, the time zone from 0:00 to 7:00. When the predetermined time zone is a 30-minute time zone from 0:00 to 7:00, 365 days × 14 = 5110 unit data sets are extracted.

[0053] FIG. 7 shows a result of plotting a plurality of (for example, 5110) extracted unit data sets 40 on a graph with the outside air temperature on the horizontal axis and the power consumption on the vertical axis.

[0054] Since base component 21 corresponds to the amount of power consumed by electrical appliances that are constantly in operation, it is highly likely to correspond to the minimum amount of power consumption at each outside temperature. Therefore, analysis unit 15 calculates base component 21 for each time period of each day included in the analysis target period according to the following steps (A1) to (A4).

[0055] Step (A1): The analysis unit 15 divides the outdoor air temperature range from the minimum outdoor air temperature Tmin to the maximum outdoor air temperature Tmax among the outdoor air temperatures indicated by the extracted multiple unit data sets 40 into multiple outdoor air temperature zones. The number of outdoor air temperature zones is determined in advance. In the example shown in FIG. 7, the analysis unit 15 divides the outdoor air temperature range from the minimum outdoor air temperature Tmin to the maximum outdoor air temperature Tmax into 30 outdoor air temperature zones. The width Δt of each of the 30 outdoor air temperature zones is the same.

[0056] Step (A2): The analysis unit 15 identifies the minimum power consumption for each outdoor air temperature zone. As shown in Fig. 7, for each of the 30 outdoor air temperature zones, the analysis unit 15 extracts a unit data set 41 indicating the minimum power consumption from among the unit data sets 40 belonging to the outdoor air temperature zone. The analysis unit 15 identifies the power consumption indicated by the unit data set 41 extracted for each of the 30 outdoor air temperature zones as the minimum power consumption for the outdoor air temperature zone.

[0057] Step (A3): The analysis unit 15 performs a regression analysis using the outside air temperature as the explanatory variable x and the minimum power consumption amount identified in step (A2) as the objective variable y to obtain a regression equation 50. The regression equation 50 is calculated using the least squares method. The graph shown in FIG. 7 depicts a regression curve 30 represented by the regression equation 50. The regression equation 50 is typically a quadratic equation (y=ax 2 +bx+c), which represents a quadratic curve as regression curve 30.

[0058] When the outside air temperature rises, for example, the amount of power consumed by the cooling operation of an air conditioner used for 24 hours increases. On the other hand, when the outside air temperature drops, for example, the amount of power consumed by the heating operation of an air conditioner used for 24 hours increases. Therefore, the regression curve 30 usually has a downward convex shape.

[0059] Procedure (A4): The analysis unit 15 determines, as the base component 21 in each time zone of each day included in the analysis target period, the value of the objective variable y obtained by substituting the outside air temperature in each time zone of each day included in the analysis target period into the explanatory variable x of the regression equation 50.

[0060] <G. Method for calculating base fixed component and base seasonal fluctuation component> Referring to FIG. 8, a specific processing method of step S6 will be described. FIG. 8 is a diagram for explaining a method for calculating the base fixed component 27 and the base seasonal fluctuation component 28.

[0061] The base fixed component 27 (see FIG. 5) corresponds to the amount of power consumed by non-season-dependent electrical equipment that is always in operation. Therefore, as shown in FIG. 8, the analysis unit 15 determines, as the base fixed component 27, the minimum value of the objective variable y (minimum power consumption) when the explanatory variable x (outside air temperature) is changed in the regression equation 50 representing the regression curve 30. In FIG. 8, the broken line 31 indicates the minimum value of the objective variable y.

[0062] The analysis unit 15 decomposes the base component 21 in each time zone of each day in the analysis target period into a base fixed component 27 and a base seasonal fluctuation component 28 other than the base fixed component 27. That is, the analysis unit 15 calculates the base seasonal fluctuation component 28 in each time zone of each day in the analysis target period by subtracting the base fixed component 27 from the base component 21 in each time zone of each day in the analysis target period.

[0063] <H. Method for calculating activity-caused fixed component> A method for calculating activity-attributable fixed component 23 will be described with reference to Fig. 9 and Fig. 10. Fig. 9 is a diagram showing a method for calculating the outside air temperature influenced component included in second component 22. Fig. 10 is a diagram showing the distribution of the difference between second component 22 and the outside air temperature influenced component.

[0064] The analysis unit 15 classifies the unit data sets 40 (see FIG. 7) for each time period on each day included in the analysis period into a first group corresponding to the heating period and a second group corresponding to the cooling period. Specifically, the analysis unit 15 classifies the unit data sets 40 corresponding to days belonging to the heating period into the first group, and classifies the unit data sets 40 corresponding to days belonging to the cooling period into the second group.

[0065] The heating season is a period when the outside temperature is lower than in the cooling season and heating is mainly used. The cooling season is a period when the outside temperature is higher than in the heating season and air conditioning is mainly used.

[0066] The heating season and the cooling season are predefined. For example, November to April is predefined as the heating season, and May to October is predefined as the cooling season.

[0067] Alternatively, the analysis unit 15 may determine the heating period and the cooling period based on the power data 12. When switching from the heating period to the cooling period and when switching from the cooling period to the heater, the amount of power consumption is low. Therefore, the analysis unit 15 uses the power data 12 to identify two months in which the average amount of power consumption per day for each month is minimal. The analysis unit 15 may determine the heating period and the cooling period with the two identified months as the boundary. Each of the two identified months is classified as either the heating period or the cooling period.

[0068] The analysis unit 15 calculates the second component 22 for each time period on each day by subtracting the base component 21 calculated for that time period on that day from the power consumption indicated by the unit data set 40 for that time period on that day included in the analysis period.

[0069] 9 shows the results of plotting points corresponding to unit data set 40 for the target time slot on each day included in the analysis period on a graph with the outside air temperature on the horizontal axis and the second component on the vertical axis. In FIG. 9, points indicated by triangles correspond to unit data set 40 for the target time slot on days that belong to the heating period. Points indicated by circles correspond to unit data set 40 for the target time slot on days that belong to the cooling period.

[0070] The analysis unit 15 calculates the activity-attributable fixed component 23 in the target time zone on the target day in accordance with the following steps (B1) to (B6).

[0071] (Step B1) If the target day belongs to the heating period, the analysis unit 15 selects multiple days that belong to the heating period from the analysis target period. If the target day belongs to the cooling period, the analysis unit 15 selects multiple days that belong to the cooling period from the analysis target period.

[0072] (Step B2) The analysis unit 15 generates correlation information indicating the correlation between the outdoor temperature in the target time slot on the selected multiple days and the second component 22. Specifically, the analysis unit 15 performs regression analysis with the outdoor temperature as the explanatory variable x' and the second component 22 as the objective variable y' to obtain regression equation 52. Regression equation 52 is calculated using the least squares method. The graph shown in FIG. 9 depicts a regression curve 32a obtained by performing regression analysis on the outdoor temperature in the target time slot on multiple days belonging to the cooling season and the second component 22. Furthermore, a regression curve 32b obtained by performing regression analysis on the outdoor temperature in the target time slot on multiple days belonging to the heating season and the second component 22 is depicted. Regression equation 52 is typically a quadratic equation (y'=a'x' 2 +b'x'+c'), and the regression curves 32a and 32b represent quadratic curves.

[0073] In the cooling season, the higher the outdoor temperature, the more power is consumed by the cooling operation of the air conditioner. Even in the cooling season, there are days when the outdoor temperature is low, and on such days, the amount of power consumed by the heating operation of the air conditioner is large. Similarly, in the heating season, the lower the outdoor temperature, the more power is consumed by the heating operation of the air conditioner. Even in the heating season, there are days when the outdoor temperature is high, and on such days, the amount of power consumed by the cooling operation of the air conditioner is large. For this reason, the regression curves 32a and 32b are usually downwardly convex.

[0074] (Step B3): The analysis unit 15 calculates the minimum value of the objective variable y' (second component 22) when the explanatory variable x' (outdoor air temperature) is changed in the regression equation 52. Fig. 9 shows a dashed line 33a indicating the minimum value of the objective variable y' calculated from the regression curve 32a corresponding to the cooling season, and a dashed line 33b indicating the minimum value of the objective variable y' calculated from the regression curve 32b corresponding to the heating season.

[0075] (Step B4): The analysis unit 15 calculates the outdoor temperature influence by subtracting the minimum value calculated in step B3 from the value of the objective variable y' obtained by substituting the outdoor temperature in the target time period of each of the selected days into the explanatory variable x' of the regression equation 52. In Fig. 9, the difference between the regression curve 32a corresponding to the cooling season and the dashed line 33a is shown as the outdoor temperature influence corresponding to the cooling season. Furthermore, the difference between the regression curve 32b corresponding to the heating season and the dashed line 33b is shown as the outdoor temperature influence corresponding to the heating season.

[0076] (Step B5): Analysis unit 15 calculates the difference between second component 22 for the target time period on each of the selected days and the outdoor temperature influence calculated in step B4. In Fig. 10, the difference between second component 22 and the outdoor temperature influence during the heating season is indicated by a triangle mark, and the difference between second component 22 and the outdoor temperature influence during the cooling season is indicated by a circle mark.

[0077] (Step B6): The analysis unit 15 calculates the activity-attributable fixed component 23 based on the difference calculated in step B5. For example, the analysis unit 15 calculates a representative value (e.g., an average value) of the difference between the second component 22 and the outside temperature influence component in the target time periods of the selected multiple days as the activity-attributable fixed component 23.

[0078] Human activities vary depending on the day of the week, whether it is a holiday, and whether it is a public holiday. Therefore, the analysis unit 15 may classify each of the selected multiple days into one of multiple groups defined by at least one of the day of the week, whether it is a holiday, and whether it is a public holiday. The analysis unit 15 may determine a representative value of the difference between the second component 22 and the outdoor temperature influence component in the target time zone of a day belonging to each of the multiple groups as the activity-attributed fixed component 23 in the target time zone of a day belonging to the group.

[0079] The multiple groups include, for example, seven groups: "Monday," "Tuesday," "Wednesday," "Thursday," "Friday," "Saturday," and "Sunday and holidays." Note that a day that is a Monday through Saturday and is also a holiday is classified into the group "Sunday and holidays."

[0080] The activity-based fixed component 23 corresponds to the amount of power consumed by non-seasonally dependent electrical appliances that are operated during human activity. Therefore, the activity-based fixed component 23 during time periods when people are not active is preferably zero. Therefore, the analysis unit 15 may adjust the activity-based fixed component 23 so that the smallest activity-based fixed component 23 among all time periods of a plurality of groups is zero.

[0081] For example, the analysis unit 15 classifies the selected plurality of days into the above seven groups, and calculates the average value of the difference between the second component 22 and the outside air temperature influence for each group and each time zone according to the above procedure B5. The analysis unit 15 determines the minimum value among the average values calculated for all time zones of the seven groups as the offset amount. The analysis unit 15 determines, as the activity-caused fixed component 23 in the target time zone of the days belonging to each of the seven groups, a value obtained by subtracting the offset amount from the average value of the difference between the second component 22 and the outside air temperature influence in the target time zone of the days belonging to the group.

[0082] In addition, when the sum of the activity-caused fixed component 23 and the base component 21 in the target time zone of the target day is greater than the power consumption amount in the target time zone of the target day, the analysis unit 15 adjusts the activity-caused fixed component 23 so that the difference between the sum and the power consumption becomes smaller.

[0083] FIG. 11 is a diagram showing an example of changes in the activity-caused fixed component 23 in a week. FIG. 11 shows the activity-caused fixed component 23 of a store that closes at night. Since it closes at night, the activity-caused fixed component 23 decreases at night and increases during the day.

[0084] <I. Method for Calculating Activity-Caused Seasonal Variation Component and Residual Component> With reference to FIGS. 12 and 13, a method for calculating the activity-caused seasonal variation component 25 will be described. FIG. 12 is a diagram showing the distribution of the fourth component 24 obtained by removing the activity-caused fixed component 23 from the second component 22 with respect to the outside air temperature. FIG. 13 is a diagram showing an example of a regression curve 34 indicating the correlation between the outside air temperature and the fourth component.

[0085] The analysis unit 15 calculates the activity-caused seasonal variation component 25 and the residual component 26 in the target time zone of the target day according to the following procedures (C1) to (C6).

[0086] (Step C1) If the target day belongs to the heating period, the analysis unit 15 selects multiple days that belong to the heating period from the analysis target period. If the target day belongs to the cooling period, the analysis unit 15 selects multiple days that belong to the cooling period from the analysis target period.

[0087] (Step C2) The analysis unit 15 calculates the fourth component of the target time period for the selected multiple days. Specifically, for each of the selected multiple days, the analysis unit 15 calculates the fourth component 24 for the target time period of that day by subtracting the activity-attributable fixed component 23 from the second component 22 for the target time period of that day. Figure 12 shows the distribution of the fourth component 24 with respect to the outside temperature for multiple days. Note that Figure 12 shows the distribution of the fourth component 24 with respect to the outside temperature not only for the target time period but for all time periods.

[0088] (Step C3) Analysis unit 15 generates correlation information indicating the correlation between the outside temperature in the target time slot on the selected multiple days and the fourth component 24. Specifically, analysis unit 15 performs regression analysis with the outside temperature as the explanatory variable x″ and the fourth component 24 as the objective variable y″ to obtain regression equation 54. Regression equation 54 is calculated using the least squares method. The graph shown in FIG. 13 depicts regression curve 34 obtained by regression analysis of the outside temperature in the target time slot and the fourth component 24. Regression equation 54 is typically a quadratic equation (y″=a″x″ 2 +b”x”+c”), which represents a quadratic curve as regression curve 34.

[0089] (Step C4): The analysis unit 15 calculates the minimum value of the objective variable y″ (fourth component 24) when the explanatory variable x″ (outdoor air temperature) is changed in the regression equation 54. FIG. 13 shows a dashed line 35 indicating the minimum value calculated from the regression curve 34.

[0090] (Procedure C5): The analysis unit 15 calculates the activity-induced seasonal variation component 25 by subtracting the minimum value calculated in Procedure C4 from the value of the objective variable y” obtained by substituting the outside air temperature in the target time zone of the target day into the explanatory variable x” of the regression equation 54. In FIG. 13, the difference between the regression curve 34 corresponding to the target time zone and the broken line 35 is shown as the activity-induced seasonal variation component 25.

[0091] The analysis unit 15 calculates the residual component 26 by subtracting the activity-induced seasonal variation component 25 from the fourth component 24 for the target time zone of the target day.

[0092] FIG. 14 is a diagram showing the variation of the residual component 26 in a part of the analysis target period. The base component 21 is calculated from the power consumption in the late-night time zone during the analysis target period. The activity-induced fixed component 23 and the activity-induced seasonal variation component 25 are calculated using the correlation between the outside air temperature and the power consumption in the target time zone of the days belonging to the same group as the target day and the outside air temperature in the target time zone of the target day. Therefore, the sum of the base component 21, the activity-induced fixed component 23, and the activity-induced seasonal variation component 25 calculated for the target time zone of the target day can be larger than the power consumption (measured value) in the target time zone of the target day. As a result, as shown in FIG. 14, the residual component 26 can take a negative value.

[0093] <J. Notification Process> The notification unit 16 causes the analysis result of the analysis unit 15 to be displayed on the display 108.

[0094] FIG. 15 is a diagram showing an example of the display of the analysis result. In FIG. 15, the time variations of each of the base fixed component 27, the base seasonal variation component 28, the activity-induced fixed component 23, and the fourth component 24 (the synthesis of the activity-induced seasonal variation component 25 and the residual component 26) decomposed from the power consumption of the customer are shown. By confirming the analysis result shown in FIG. 15, the power supplier can consider the time zone for energy saving, the components to be reduced, etc., and can present the consideration result to the customer. Alternatively, the analysis result may be presented to the customer.

[0095] FIG. 16 is a diagram showing another example of display of analysis results. FIG. 16 shows a duration curve in which the amount of power consumption for each time period in the analysis period is arranged in descending order. The electricity rate is determined according to the maximum demand power (demand value). The demand value corresponds to the amount of power consumption at the leftmost side of the duration curve shown in FIG. 16. In the example shown in FIG. 16, the demand value increases with an increase in the fourth component 24, which is mainly occupied by the activity-attributable seasonal fluctuation component 25. Therefore, by checking the duration curve shown in FIG. 16, the electricity supplier can suggest to the consumer that the electricity rate can be reduced by, for example, adjusting the temperature setting of the air conditioner.

[0096] 15 and 16, the activity-related fixed component 23 accounts for a large proportion of the power consumption. The activity-related fixed component 23 corresponds to the amount of power consumed due to human activity, and therefore can have a significant impact on energy conservation.

[0097] FIG. 17 is a diagram showing an example of changes in the activity-induced fixed components 23 of a plurality of offices. FIG. 18 is a diagram showing an example of changes in the activity-induced fixed components 23 of a plurality of general supermarkets. FIG. 19 is a diagram showing an example of changes in the activity-induced fixed components 23 of a plurality of accommodation facilities. As shown in FIG. 17 to FIG. 19, the activity-induced fixed components 23 can change significantly depending on the type of business. Furthermore, the activity-induced fixed components 23 can change significantly depending on the business form even in the same business form. This means that energy can be saved by changing the business form. Therefore, the notification unit 16 may notify the result of a comparison between the activity-induced fixed components 23 of the consumer and other consumers in the same business form. Comparison with the activity-induced fixed components 23 of other consumers in the same business form helps in considering measures for energy saving.

[0098] When an energy-saving action is taken more efficiently than usual, the residual component 26 becomes a large negative value. Therefore, the notification unit 16 may evaluate the energy-saving action taken by the consumer according to the result of comparing the residual component 26 with a negative threshold value, and notify the consumer of the evaluation result as information related to energy saving. This allows the consumer to recognize the effect of the energy-saving action being taken.

[0099] Conversely, when the amount of wasted power increases more than usual, the residue component 26 takes on a large positive value. Therefore, the notification unit 16 may detect an abnormality in the amount of power consumption based on the result of comparing the residue component 26 with a positive threshold, and notify the consumer of the evaluation result as information related to energy saving. This allows the consumer to consider taking energy saving actions. Below, a method of detecting an abnormality and a method of identifying the cause of the abnormality based on the residue component 26 will be described.

[0100] (J-1. Method of detecting abnormalities) As described above, the residue components 26 fluctuate due to the use of the electrical device 220 that is different from the normal one. Therefore, the notification unit 16 detects an abnormality based on the analysis results of the residue components 26 in a detection period consisting of one or multiple consecutive time periods. The detection period is, for example, 30 minutes, 3 hours, or 1 day. When the length of the time period is 30 minutes and the length of the detection period is 30 minutes, the detection period includes one time period. When the length of the time period is 3 hours and the length of the detection period is 3 hours, the detection period includes six time periods. When the length of the time period is 30 minutes and the detection period is 1 day, the detection period includes 48 time periods. The notification unit 16 may detect an abnormality based on the analysis results of the residue components 26 in multiple detection periods having different time lengths, or may detect an abnormality based on the analysis results of the residue components 26 in one detection period.

[0101] The notification unit 16 detects an abnormality based on the analysis results of the residue components 26 during the detection period according to at least one of a first pattern and a second pattern described below.

[0102] (First pattern) When the electrical device 220 is used normally, the residual components 26 during the detection period are likely to be influenced by the residual components 26 during a certain period immediately preceding the detection period. Therefore, the notification unit 16 creates an autoregressive model using data on the residual components 26 for each unit time included in a learning period selected from the analysis period. The unit time has the same length as the detection period. The learning period is, for example, four weeks. However, the length of the learning period is not limited to four weeks and can be set as appropriate. The autoregressive model is created using a known method.

[0103] In addition, the shorter the detection target period, the more likely the residue components 26 in the detection target period are to be influenced by the residue components 26 in a certain period immediately preceding the detection target period. Therefore, when the detection target period is short (e.g., 30 minutes), it is preferable that the notification unit 16 detects an abnormality using the first pattern.

[0104] Fig. 20 is a diagram for explaining an autoregressive model. The autoregressive model illustrated in Fig. 20 is a model that estimates a predicted value of the residue component 26 at unit time t using data of analysis results of the residue component 26 at unit times t-1, t-2, ..., tr that are past the unit time t that is the detection target period. Note that the unit time k is the unit time next to the unit time k-1.

[0105] An autoregressive model is typically expressed by the following equation (1). x t =a1x t-1 +a2x t-2 +···+a r x t-r Formula (1) In formula (1), x k (k is an integer from t-1 to tr) represents a representative value of the analysis results of the residue components 26 calculated for one or more time periods included in the unit time k. The representative value is, for example, an integrated value, an average value, a median value, etc. When only one time period is included in the unit time, the representative value may be the value of the analysis results of the residue components 26 calculated for that time period. Note that when the length of the unit time is the same as the length of the time period, xk represents the analysis result of the residual component 26 at unit time k. t represents the predicted value of the residual component 26 at unit time t. r is determined using data from the training period and known methods such as the Yule-Walker method, the least squares method, the maximum likelihood method, or the Burg method.

[0106] r is an integer equal to or greater than 2. When the length of the unit time is 30 minutes, r is set to, for example, any one of 48, 12, 6, and 2. When r=48, a predicted value of the residual component 26 at unit time t is estimated from the analysis results of the residual component 26 at unit times t-1 to t-48 for the past day. Note that r may be determined according to the evaluation results of an autoregressive model. For the evaluation of the autoregressive model, a known information criterion such as AIC (Akaike's Information Criterion) or BIC (Bayesian Information Criterion) is used.

[0107] As described above, the analysis result of the residue component 26 may be negative. Therefore, the predicted value of the residue component 26 at unit time t may also be negative. For example, if the analysis result of the residue component 26 at unit time t-1 prior to unit time t is negative, the predicted value of the residue component 26 at unit time t may be negative due to the influence of the analysis result.

[0108] When an amount of power that is significantly different from the average power consumption at the outside temperature in the unit time t in a group to which the day belongs is consumed in the unit time t, the difference between the predicted value of the residue component 26 in the unit time t and the representative value of the analysis result of the residue component 26 in the unit time t becomes large. Therefore, the notification unit 16 compares the difference between the representative value of the analysis result of the residue component 26 in the unit time t, which is the detection target period, and the predicted value of the residue component 26 in the unit time t with a threshold value Th, and detects an abnormality in the power consumption in the detection target period according to the comparison result. The threshold value Th is determined, for example, according to the distribution (e.g., probability distribution) of the difference between the representative value of the analysis result of the residue component 26 for each unit time and the predicted value of the residue component 26 in the unit time predicted using an autoregressive model.

[0109] 21 is a diagram showing an example of a probability distribution of the difference between the representative value and the predicted value of the analysis result of the residue component 26 for each unit time. The notification unit 16 may create a probability distribution of the difference between the representative value and the predicted value of the analysis result of the residue component 26 for each unit time, for example, by using a sample distribution obtained from the difference between the representative value and the predicted value of the analysis result of the residue component 26 for each unit time. The sample distribution may be created from all data during the analysis period, or may be created from data for a plurality of unit times randomly selected from the analysis period.

[0110] As the threshold value Th, a difference value between the representative value of the analysis result of the residue component 26 having an upper probability of Pu% and the predicted value (hereinafter referred to as "threshold Th1") and a difference value between the representative value of the analysis result of the residue component 26 having a lower probability of Pd% and the predicted value (hereinafter referred to as "threshold Th2"). Pu% and Pd% are, for example, 10%. However, the values ​​of Pu% and Pd% are not particularly limited. The notification unit 16 determines that the power consumption in the unit time t is normal when the difference between the representative value of the analysis result of the residue component 26 in the unit time t and the predicted value is equal to or greater than the threshold Th2 and equal to or less than the threshold Th1. The notification unit 16 determines that the power consumption in the unit time t is abnormal when the difference between the representative value of the analysis result of the residue component 26 in the unit time t and the predicted value is less than the threshold Th2 or greater than the threshold Th1.

[0111] When the representative value of the analysis result of the residue components 26 is smaller than the predicted value, it is expected that energy saving is being implemented. Therefore, as the threshold value Th, only the difference value (threshold value Th1) between the representative value of the analysis result of the residue components 26 with the upper probability of Pu% and the predicted value may be set. In this case, the notification unit 16 determines that the power consumption in the unit time t is normal when the representative value of the analysis result of the residue components 26 in the unit time t is smaller than the predicted value, or when the difference between the representative value of the analysis result of the residue components 26 in the unit time t and the predicted value is equal to or smaller than the threshold value Th1. The notification unit 16 determines that the power consumption in the unit time t is abnormal when the representative value of the analysis result of the residue components 26 in the unit time t is larger than the predicted value and the difference between the representative value and the predicted value is larger than the threshold value Th1.

[0112] In this way, the notification unit 16 detects an abnormality in the power consumption during the detection period based on the comparison result between the difference between the representative value and the predicted value of the analysis result of the residue components 26 during the unit time t, which is the detection period, and the threshold value Th.

[0113] FIG. 22 is a diagram showing the component analysis results of the power consumption and the predicted value of the residual component 26. FIG. 22 shows the component analysis results of the power consumption for each of 48 time periods on August 12. That is, FIG. 22 shows an example in which the number of time periods included in the unit time is one (the length of the unit time is 30 minutes). The base fixed component 27, the base seasonal variation component 28, the activity-attributed fixed component 23, the activity-attributed seasonal variation component 25, and the residual component 26 are shown by stacked bar graphs. However, the analysis result of the residual component 26 may also take a negative value. Therefore, the analysis result of the residual component 26 is drawn shifted to the right of the other components. If the residual component 26 extends downward from the position of the upper end of the activity-attributed seasonal variation component 25, it indicates that the analysis result of the residual component 26 is a negative value. If the analysis result of the residual component 26 extends upward from the position of the upper end of the activity-attributed seasonal variation component 25, it indicates that the analysis result of the residual component 26 is a positive value.

[0114] 22 shows the change in the total value of the power consumption excluding the residual component 26 and the predicted value of the residual component 26. The length of the error bar drawn on the line graph is the threshold value Th1. Therefore, the error bar shows the normal range of the residual component 26.

[0115] 22, the difference between the analysis result of the residue component 26 and the predicted value exceeds the upper limit of the error bar (i.e., the threshold value Th1) in three time periods enclosed by the frame line 60. Therefore, the notification unit 16 detects an abnormality in the power consumption amount for the three time periods.

[0116] (Second pattern) The second pattern is a method that does not use an autoregressive model, unlike the first pattern. The longer the length of the detection target period, the less the residual components 26 in the detection target period are influenced by the residual components 26 in a certain period immediately preceding the detection target period. Therefore, it is preferable that the notification unit 16 detects an abnormality using the second pattern when the length of the detection target period is long (for example, one day).

[0117] When an amount of power consumed during a detection period is significantly different from the average amount of power consumption per unit time in a group to which a day including the detection period belongs, the representative value of the analysis result of the residue components 26 during the detection period significantly deviates from the representative values ​​of the analysis result of the residue components 26 during other unit times. Therefore, the notification unit 16 compares the representative value of the analysis result of the residue components 26 during the unit time t, which is the detection period, with a threshold value Th, and detects an abnormality in the amount of power consumption during the detection period according to the comparison result. The threshold value Th is determined, for example, according to the distribution (e.g., probability distribution) of the representative value of the analysis result of the residue components 26 for each unit time included in the analysis period.

[0118] 23 is a diagram showing an example of a probability distribution of the representative value of the residue component 26 in a unit time. The notification unit 16 may create a probability distribution of the representative value of the residue component 26 in a unit time, for example, by using a sample distribution obtained from a representative value of an analysis result of the residue component 26 in a unit time during an analysis target period. Specifically, the probability distribution of the representative value of the residue component 26 in a unit time of a certain day is created from a sample distribution of data of the unit time of a day belonging to the same group as the day in question.

[0119] As the threshold value Th, a value of the residue component 26 where the upper probability is Pu% (hereinafter referred to as "threshold Th1'") and a value of the residue component 26 where the lower probability is Pd% (hereinafter referred to as "threshold Th2'") are set. Pu% and Pd% are, for example, 10%. However, the values ​​of Pu% and Pd% are not particularly limited. The notification unit 16 determines that the power consumption in the unit time t is normal when the representative value of the analysis result of the residue component 26 in the unit time t is equal to or greater than the threshold Th2' and equal to or less than the threshold Th1'. The notification unit 16 determines that the power consumption in the unit time t is abnormal when the representative value of the analysis result of the residue component 26 in the unit time t is less than the threshold Th2' or greater than the threshold Th1'.

[0120] As in the first pattern, only the value of the residue components 26 at which the upper probability is Pu% (threshold Th1') may be set as the threshold Th. In this case, the notification unit 16 determines that the power consumption in the unit time t is normal when the representative value of the analysis result of the residue components 26 in the unit time t is equal to or smaller than the threshold Th1'. The notification unit 16 determines that the power consumption in the unit time t is abnormal when the representative value of the analysis result of the residue components 26 in the unit time t is greater than the threshold Th1'.

[0121] In this manner, the notification unit 16 detects an abnormality in the amount of power consumption during the detection period based on the comparison result between the representative value of the analysis result of the residue components 26 during the detection period and the threshold value Th.

[0122] FIG. 24 is a diagram showing a representative value of the analysis result of the residue component 26 for each unit time (one day) and a threshold value Th1'. FIG. 24 shows the representative value of the analysis result of the residue component 26 for each unit time from September 13th to September 21st. In the example shown in FIG. 24, the representative value is an integrated value of the analysis result of the residue component 26 for each unit time. The error bars drawn in FIG. 24 indicate the range of the residue component 26 where the lower probability of the probability distribution is 90%. Therefore, the error bars indicate the normal range of the residue component 26.

[0123] 24, the representative value of the analysis result of the residue components 26 exceeds the upper limit of the error bar (i.e., threshold value Th1') in the area surrounded by frame line 60. Therefore, notification unit 16 detects an abnormality in the amount of power consumption for the unit time (September 19, 2016) corresponding to that area.

[0124] (J-2. Identifying the cause of the abnormality) When the notification unit 16 detects an abnormality according to the above (J-1. Method of detecting an abnormality), it identifies the cause of the abnormality and notifies the user of the identified cause.

[0125] Fig. 25 is a diagram showing phenomena that cause an increase in the residual component 26 and a method for determining the presence or absence of the phenomenon. As shown in Fig. 25, the following (A) to (D) are assumed as phenomena that cause an increase in the residual component 26. (A) Forgetting to turn off electrical equipment and installing new electrical equipment that is operated constantly. (B) Operate electrical equipment on holidays in response to human activity. (C) Changes in the operating time of electrical equipment operated during human activity. (D) Excessive power consumption of seasonally dependent electrical appliances that operate during human activity.

[0126] If an electric appliance operated during human activity is forgotten to be turned off, the residual component 26 increases throughout the day. Alternatively, the residual component 26 increases throughout the day immediately after a constantly operating electric appliance is newly installed. Therefore, the notification unit 16 compares the minimum value of the residual component 26 in the entire time period (from 0:00 to 24:00) including the time period in which the abnormality was detected with the threshold value Tha. The threshold value Tha is, for example, Xa% of the analysis value of the base fixed component 27. The value of Xa is not particularly limited and is determined in advance. If the minimum value of the residual component 26 is greater than the threshold value Tha, the notification unit 16 detects the occurrence of an abnormality caused by forgetting to turn off the electric appliance or the installation of a constantly operating electric appliance, and notifies the user of the cause.

[0127] When an electrical appliance is operated in response to human activity on a holiday, the residual component 26 becomes large during the time period during which the electrical appliance is active. Therefore, the integrated value or average value of the residual component 26 for one day becomes large. Therefore, the notification unit 16 compares the integrated value or average value of the analysis result of the residual component 26 for all time periods of the day (from 0:00 to 24:00) including the time period during which the abnormality was detected with the threshold value Thb. The threshold value Thb is, for example, a value obtained by adding Xb% of the integrated value or average value of the activity-attributable fixed component 23 for all time periods of the day to the total value of the predicted values ​​of the residual component 26 for all time periods of the day. The value of Xb is not particularly limited and is determined in advance. When the integrated value or average value of the residual component 26 is larger than the threshold value Thb, the notification unit 16 detects the occurrence of an abnormality caused by the operation of the electrical appliance in response to human activity on a holiday and notifies the cause.

[0128] Fig. 26 is a diagram showing the integrated values ​​of the analysis results of the residual components 26 for each day from November 10, 2016 to December 31, 2016. In Fig. 26, the total value of the predicted values ​​of the residual components 26 for all time periods for each day is shown by a line graph. The error bars drawn superimposed on the line graph have a length of Xb% (=80%) of the integrated value of the activity-attributable fixed components 23 for all time periods.

[0129] As shown by the frame line 61 in Fig. 26, the integrated value of the residue component 26 on December 23rd exceeds the range indicated by the error bar. This is because electrical appliances were operated in response to human activity on December 23rd, which is a holiday.

[0130] FIG. 27 is a diagram showing the component analysis results of the power consumption amount on December 23, 2016. As described above, the activity-attributable fixed component 23 is determined based on the average value of the difference between the second component 22 and the outdoor temperature influence component in the target time slot of the day belonging to each of the seven groups. Since December 23 is a public holiday, the activity-attributable fixed component 23 for each time slot on December 23 is determined based on the average value of the difference between the second component 22 and the outdoor temperature influence component in the relevant time slot of the day belonging to the group "Sunday and public holidays". Therefore, as shown in FIG. 27, the activity-attributable fixed component 23 for each time slot is small. However, since the electric appliances were operated in response to human activities despite it being a public holiday, the amount of power consumed by the electric appliances appears as the residual component 26. As a result, the analysis result of the residual component 26 is large.

[0131] As shown in Figures 26 and 27, since the integrated value of the analysis results of the residue components 26 is greater than the threshold value Thb, the notification unit 16 detects the occurrence of an abnormality caused by the operation of electrical equipment in response to human activity on December 23, which is a holiday, and notifies the user of the cause.

[0132] Returning to FIG. 25, when the operating time of an electrical appliance operated during human activity is changed, the residual component 26 of the time period in which the electrical appliance is changed from not operating to operating becomes large. Therefore, the notification unit 16 compares the analysis result of the residual component 26 of the time period in which the abnormality is detected with a threshold value Thc. The threshold value Thc is, for example, the sum of the predicted value of the residual component 26 of the time period in which the abnormality is detected and Xc% of the activity-attributable fixed component 23 of the time period. The value of Xc is not particularly limited and is determined in advance. When the analysis value of the residual component 26 is larger than the threshold value Thc, the notification unit 16 detects the occurrence of an abnormality caused by the change in the operating time of the electrical appliance operated during human activity and notifies the cause.

[0133] Fig. 28 is a diagram showing the component analysis results of the power consumption amount and the threshold value Thc. Fig. 28 shows the component analysis results of the power consumption amount for each of the 48 time periods on August 12th.

[0134] The line graph in Fig. 28 shows the change in the total value of the power consumption excluding the residual component 26 and the predicted value of the residual component 26. The error bar drawn over the line graph has a length corresponding to the threshold value Thc (30% of the activity-induced fixed component 23 in Fig. 28) and extends upward from the line graph.

[0135] In the example shown in Fig. 28, the analysis value of the residue components 26 is greater than the upper limit of the error bar (i.e., the threshold value Thc) during the time period surrounded by the frame line 62. This is because the electrical appliance operated during human activity was started earlier than usual. In response to the analysis result of the residue components 26 being greater than the threshold value Thc, the notification unit 16 detects the occurrence of an abnormality caused by a change in the operating time of the electrical appliance operated during human activity, and notifies the user of the cause.

[0136] Returning to FIG. 25, when the power consumption of a seasonally-dependent electrical appliance operated during human activity is excessively large, the degree of change in the analysis results of the residue components 26 in response to changes in the outside air temperature (hereinafter referred to as "temperature sensitivity") increases. Therefore, the notification unit 16 compares the temperature sensitivity of the residue components 26 in a specific time period in a certain period with the threshold value Thd. The certain period is, for example, one month, and is determined in advance. The specific time period includes one or more time periods determined in advance. The specific time period is set to a time period during which people are active. For example, the specific time period is a time period from 12:00 to 15:00.

[0137] For example, the notification unit 16 calculates the increase in the residue components 26 when the outside air temperature rises by ΔT°C from the relationship between the analysis results of the residue components 26 in a specific time period in a certain period and the outside air temperature as the temperature sensitivity of the residue components 26. ΔT°C is, for example, 5°C.

[0138] The threshold value Thd is, for example, Xd% of the temperature sensitivity of the activity-induced seasonal variation component 25 in a specific time period within a certain period. The notification unit 16 calculates, from the relationship between the activity-induced seasonal variation component 25 and the outside air temperature in a certain period, the increase in the activity-induced seasonal variation component 25 when the outside air temperature rises by ΔT °C as the temperature sensitivity of the activity-induced seasonal variation component 25. ΔT °C is, for example, 5 °C. The value of Xd is not particularly limited and is determined in advance.

[0139] When the temperature sensitivity of the residue component 26 is greater than the threshold value Thd, the notification unit 16 detects the occurrence of an abnormality caused by an excessive power consumption of a season-dependent electrical device that operates during human activities and notifies the cause.

[0140] FIG. 29 is a diagram showing the temperature sensitivity of the residue component 26. In FIG. 29, for a specific time period (the time period from 12:00 to 15:00) within a certain period in summer, the average value of the analysis results of the residue component 26 for each temperature zone is shown. The width of the temperature zone is 5 °C. In FIG. 29, the temperature sensitivity of the residue component 26 is shown by a solid line, and the threshold value Thd is shown by a broken line. In the graph shown in FIG. 29, the horizontal axis represents the temperature zone, and the vertical axis represents the average value of the residue component 26. Therefore, the temperature sensitivity of the residue component 26 corresponds to the slope of the graph.

[0141] In the example shown in FIG. 29, the increase in the average value of the residue component 26 when changing from the temperature zone of 30 to 35 °C to the temperature zone of 35 to 40 °C is greater than the threshold value Thd. This is because a season-dependent electrical device (for example, an air conditioner) that operates during human activities is operated excessively. Therefore, when the temperature sensitivity of the residue component 26 is greater than the threshold value Thd, the notification unit 16 detects the occurrence of an abnormality caused by an excessive power consumption of a season-dependent electrical device that operates during human activities and notifies the cause.

[0142] <K. Modification Example> In the above description, the analysis unit 15 calculates the outdoor temperature influenced component of the second component 22 for the target time period according to steps (B1) to (B4), and calculates the activity-attributable fixed component 23 based on the difference between the second component 22 for the target time period and the outdoor temperature influenced component. However, the analysis unit 15 may determine the minimum value of the objective variable y' (second component 22) when the explanatory variable x' (outdoor temperature) is changed in the regression equation 52 obtained by step B2 as the activity-attributable fixed component 23 for the target time period.

[0143] In the above description, the process according to (J-2. Identifying the cause of anomaly) is performed after the process according to (J-1. Method for detecting anomaly). However, (J-1. Method for detecting anomaly) may be omitted. In this case, the notification unit 16 may detect the occurrence of an anomaly caused by forgetting to turn off an electric appliance that is constantly operated or the installation of a new electric appliance that is constantly operated, in response to the minimum value of the residue components 26 being greater than the threshold value Tha, and notify the occurrence of the anomaly. Alternatively, the notification unit 16 may detect the occurrence of an anomaly caused by operating an electric appliance in response to human activity on a holiday, in response to the integrated value or average value of the residue components 26 being greater than the threshold value Thb, and notify the occurrence of the anomaly. Alternatively, the notification unit 16 may detect the occurrence of an anomaly caused by changing the operating time of an electric appliance that is operated during human activity, in response to the analysis value of the residue components 26 being greater than the threshold value Thc, and notify the occurrence of the anomaly. Alternatively, when the temperature sensitivity of the residue component 26 is greater than the threshold value Thd, the notification unit 16 detects the occurrence of an abnormality caused by excessive power consumption of seasonally-dependent electrical equipment that is operated during human activity, and notifies the occurrence of the abnormality.

[0144] In the example shown in Fig. 1, the smart meter 210, the server device 300, and the analysis device 100 exchange data with each other via a network. However, the method of exchanging data among the smart meter 210, the server device 300, and the analysis device 100 is not limited to the example shown in Fig. 1. For example, the server device 300 may directly acquire power energy data from the smart meter 210 without going through a network. Furthermore, the analysis device 100 may be connected to the server device 300 without going through a network, and directly acquire power energy data of the building 200 from the server device 300.

[0145] In the above description, "greater than or equal to" may be replaced with "greater than". "Less than or equal to" may be replaced with "less than". Similarly, "greater than" may be replaced with "greater than or equal to". "Less than" may be replaced with "less than or equal to".

[0146] <L.まとめ> As described above, the analysis device 100 includes the first acquisition unit 11, the second acquisition unit 13, and the analysis unit 15. The first acquisition unit 11 acquires power data 12 indicating the power consumption of the consumer measured for each time period during the analysis period. The second acquisition unit 13 acquires outdoor temperature data 14 indicating the outdoor temperature for each time period in the area of ​​the consumer during the analysis period. The analysis unit 15 analyzes the power consumption for each time period on each day included in the analysis period. The analysis unit 15 uses the power data 12 and the outdoor temperature data 14 to generate first correlation information indicating a correlation between the outdoor temperature and the minimum power consumption at the outdoor temperature. The analysis unit 15 uses the first correlation information and the outdoor temperature data 14 to decompose the power consumption for the target time period on each day included in the analysis period into a base component 21 that does not depend on human activity and a second component 22 other than the base component 21. The analysis unit 15 uses the outside temperature data 14 to generate second correlation information indicating a correlation between the outside temperature in a target time period and the second component 22 for multiple days included in the analysis period. The analysis unit 15 uses the second correlation information and the outside temperature data 14 to decompose the second component 22 in the target time period for each of the multiple days into an activity-attributed fixed component 23 that does not vary according to the outside temperature, and a fourth component 24 other than the activity-attributed fixed component 23.

[0147] According to the above configuration, unlike the technology described in Non-Patent Document 1, it is possible to decompose the power consumption into a base component 21, an activity-attributable fixed component 23, and a fourth component 24 without assuming that there are days when the air conditioning equipment is not operated. Therefore, it is possible to analyze the power consumption of various consumers.

[0148] The first correlation information indicates a correlation between an outside temperature and a minimum power consumption amount at the outside temperature. The minimum power consumption amount is expected to be the amount of power consumed by an electrical device that is constantly operating, regardless of human activity. Therefore, by using the first correlation information, the amount of power consumed by an electrical device that is constantly operating and is not caused by human activity can be easily calculated as the base component 21.

[0149] The second correlation information indicates the correlation between the outside air temperature in the target time period and the second component 22. The second component 22 is a component other than the base component 21, and therefore corresponds to the amount of electricity consumed due to human activity. By using the outside air temperature data 14, the second component 22 can be easily decomposed into an activity-attributable fixed component 23 that does not vary according to the outside air temperature, and a fourth component 24 other than the activity-attributable fixed component 23. The activity-attributable fixed component 23 corresponds to the amount of electricity consumed by non-season-dependent electrical appliances operated during human activity. The fourth component 24 is the remaining component, and mainly includes the amount of electricity consumed by season-dependent electrical appliances operated during human activity.

[0150] The embodiments disclosed herein should be considered to be illustrative and not restrictive in all respects. The scope of the present invention is defined by the claims, not the above description, and is intended to include all modifications within the meaning and scope of the claims. [Explanation of symbols]

[0151] 10 memory unit, 11 first acquisition unit, 12 power amount data, 13 second acquisition unit, 14 outdoor temperature data, 15 analysis unit, 16 notification unit, 20 power consumption, 21 base component, 22 second component, 23 activity-induced fixed component, 24 fourth component, 25 activity-induced seasonal variation component, 26 residual component, 27 base fixed component, 28 base seasonal variation component, 30, 32a, 32b, 34 regression curve, 31, 33a, 33b, 35 dashed line, 40, 41 unit data set, 50, 52, 54 regression equation, 60 to 62 frame line, 100 analysis device, 101 processor, 102 ROM, 103 RAM, 105 communication IF 106 operation key, 107 power supply circuit, 108 display, 110 program, 200 building, 210 smart meter, 220 Electrical equipment, 300 server equipment, SYS system.

Claims

1. 1. An analytical apparatus comprising: A first acquisition unit that acquires power amount data indicating power consumption of a consumer measured for each time period during an analysis target period; A second acquisition unit that acquires outdoor temperature data indicating outdoor temperatures for each time period in the area of ​​the consumer during the analysis period; An analysis unit that analyzes the amount of power consumption in each time period of each day included in the analysis period, The analysis unit includes: generating first correlation information indicating a correlation between the outside air temperature and a minimum power consumption amount at the outside air temperature using the power amount data and the outside air temperature data; using the first correlation information and the outside air temperature data, decomposing the amount of power consumption in a target time period of each day included in the analysis period into a first component corresponding to an amount of power consumed by an electrical appliance that is constantly operated and a second component obtained by subtracting the first component from the amount of power consumption; using the outside temperature data, generate second correlation information indicating a correlation between the outside temperature during the target time period and the second component for a plurality of days included in the analysis target period; an analysis device that uses the second correlation information and the outside temperature data to decompose the second component in the target time period for each of the plurality of days into a third component corresponding to the amount of electricity consumed by a non-seasonally dependent electrical appliance operated during human activity, and a fourth component obtained by subtracting the third component from the second component.

2. The second correlation information indicates a first equation obtained by a regression analysis in which the outside air temperature is a first explanatory variable and the second component is a first objective variable, The analysis unit includes: calculating an outside air temperature influence by subtracting a minimum value of the first objective variable when the first explanatory variable is changed in the first formula from a value of the first objective variable obtained by substituting the outside air temperature during the target time period of each of the plurality of days into the first explanatory variable in the first formula; Calculating a difference between the second component and the outside air temperature influence component for the target time period of each of the plurality of days; The analyzer according to claim 1 , further comprising: a detector configured to detect the third component based on the difference.

3. The analysis unit includes: classifying each of the plurality of days into one of a plurality of groups defined by at least one of a day of the week, whether it is a holiday, and whether it is a public holiday; The analysis device according to claim 2 , wherein a representative value of the differences on days belonging to each of the plurality of groups is determined as the third component in the target time period on the days belonging to the group.

4. The second correlation information indicates a first equation obtained by a regression analysis in which the outside air temperature is a first explanatory variable and the second component is a first objective variable, The analysis device according to claim 1 , wherein the analysis unit determines, as the third component, a minimum value of the first response variable when the first explanatory variable is changed in the first equation.

5. The analysis unit divides the analysis period into a first period and a second period in which the outside air temperature is lower than that of the first period, The analysis device according to claim 1 , wherein the plurality of days belong to one of the first period and the second period.

6. The analysis unit includes: Using the outside temperature data, generate third correlation information indicating a correlation between the outside temperature during the target time period and the fourth component for the plurality of days; using the third correlation information, decomposing the fourth component in the target time zone of each of the plurality of days into a fifth component and a sixth component obtained by subtracting the fifth component from the fourth component; The third correlation information indicates a third equation obtained by a regression analysis in which the outside air temperature is a second explanatory variable and the fourth component is a second objective variable, 6. The analysis device according to claim 1, wherein the analysis unit calculates the fifth component by subtracting a minimum value of the second objective variable when the second explanatory variable is changed in the third equation from a value of the second objective variable obtained by substituting the outside air temperature during the target time period on each of the plurality of days for the second explanatory variable in the third equation.

7. The analyzer according to claim 6 , further comprising a notification unit that notifies of an abnormality based on the sixth component.

8. The notification unit is predicting the predicted value using a model that predicts a predicted value of the sixth component in a detection target period, the prediction value being composed of one or a plurality of consecutive time periods, based on an analysis result of the sixth component in a time period prior to the detection target period; detecting the occurrence of the abnormality when a difference between a representative value of an analysis result of the sixth component during the detection period and the predicted value is greater than a threshold value; 8. The analysis device according to claim 7, wherein the threshold is determined according to a distribution of differences between a representative value of an analysis result of the sixth component for each unit time of the same time length as the detection period, which is included in the analysis period, and a predicted value of the sixth component for the unit time predicted using the model.

9. The notification unit detects the occurrence of the abnormality in response to a representative value of an analysis result of the sixth component in a detection period consisting of one or a plurality of consecutive time periods being greater than a threshold value; The analysis device according to claim 7 , wherein the threshold value is determined according to a distribution of a representative value of the analysis result of the sixth component for each unit time that is included in the analysis period and has the same time length as the detection period.

10. The analysis device according to claim 7 , wherein the notification unit detects the occurrence of the abnormality when a minimum value of the analysis result of the sixth component in all time periods on a target day is greater than a threshold determined from the first component.

11. The analysis device according to claim 7 , wherein the notification unit detects the occurrence of the abnormality when an integrated value or average value of the analysis results of the sixth component for all time periods on a target day is greater than a threshold value determined from the third component.

12. The analysis device according to claim 7 , wherein the notification unit detects the occurrence of the abnormality in response to an analysis result of the sixth component in the target time period on a target day being greater than a threshold determined from the third component.

13. 8. The analysis device according to claim 7, wherein the notification unit detects the occurrence of the abnormality in response to a degree of change in the analysis result of the sixth component relative to a change in outside air temperature during one or more predetermined time periods on a day belonging to a period of at least two days in length being greater than a threshold value determined from the fifth component.

14. the first correlation information indicates a fifth equation obtained by regression analysis in which an outside air temperature is a third explanatory variable and the minimum power consumption amount is a third objective variable; 14. The analysis device according to claim 1, wherein the analysis unit determines, as the first component in the target time period of each of the plurality of days, a value of the third objective variable obtained by substituting the outside air temperature in the target time period of each of the plurality of days into the third explanatory variable in the fifth equation.

15. The analysis unit includes: determining a minimum value of the third objective variable when the third explanatory variable is changed in the fifth equation as a seventh component; 15. The analysis device according to claim 14, wherein the first component in the target time period of each day included in the analysis period is decomposed into the seventh component and an eighth component obtained by subtracting the seventh component from the first component.

16. 1. A method of analysis comprising: A step of acquiring power amount data indicating power consumption of a consumer measured for each time period during an analysis period; Acquiring outdoor temperature data indicating outdoor temperatures for each time period in the area of ​​the consumer during the analysis period; Analyzing the amount of power consumption in each time period of each day included in the analysis period; The analyzing step includes: generating first correlation information indicating a correlation between the outside air temperature and a minimum power consumption amount at the outside air temperature, using the power amount data and the outside air temperature data; using the first correlation information and the outside air temperature data, decomposing the amount of power consumption in a target time period of each day included in the analysis period into a first component corresponding to an amount of power consumed by an electrical appliance that is constantly in operation and a second component obtained by subtracting the first component from the amount of power consumption; generating second correlation information indicating a correlation between the second component and the outside temperature during the target time period for a plurality of days included in the analysis target period using the outside temperature data; and using the second correlation information and the outside air temperature data, decomposing the second component into a third component corresponding to an amount of electricity consumed by a non-seasonally dependent electrical appliance operated during human activity, and a fourth component obtained by subtracting the third component from the second component.

17. A program for causing a computer to execute an analysis method, The analysis method includes: A step of acquiring power amount data indicating power consumption of a consumer measured for each time period during an analysis period; Acquiring outdoor temperature data indicating outdoor temperatures for each time period in the area of ​​the consumer during the analysis period; Analyzing the amount of power consumption in each time period of each day included in the analysis period; The analyzing step includes: generating first correlation information indicating a correlation between the outside air temperature and a minimum power consumption amount at the outside air temperature, using the power amount data and the outside air temperature data; using the first correlation information and the outside air temperature data, decomposing the amount of power consumption in a target time period of each day included in the analysis period into a first component corresponding to an amount of power consumed by an electrical appliance that is constantly in operation and a second component obtained by subtracting the first component from the amount of power consumption; generating second correlation information indicating a correlation between the second component and the outside temperature during the target time period for a plurality of days included in the analysis target period using the outside temperature data; using the second correlation information and the outside air temperature data, decomposing the second component into a third component corresponding to an amount of electricity consumed by a non-seasonally dependent electrical appliance operated during human activity, and a fourth component obtained by subtracting the third component from the second component.

Citation Information

Patent Citations

  • Method and apparatus for analysis of load kind as well as method, apparatus and system for analysis of consumption disposition as well as recording medium

    JP2000162253A

  • Device, method, and program for dividing use of energy

    JP2011108173A

  • Energy attribute determination unit and energy attribute determination method

    JP2011193602A

  • Power consumption application analysis device

    JP2017207887A

  • Power consumption volume usage decomposition system and program

    JP2020035016A