JP2017045387A - Heterogeneous energy consumption estimation device and heterogeneous characteristic estimation device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a heterogeneous energy consumption estimation device capable of estimating a heterogeneous energy consumption volume from a certain energy consumption volume with practical accuracy.SOLUTION: The heterogeneous energy consumption estimation device comprises: storage means 2 for storing sample data having a pair of first energy consumption volume in a first period and a second energy consumption volume in a second period which is the same with or different from the first period, for each consumer, as a sample value; and first threshold calculation means 3 for calculating at least one first threshold which is a boundary value sectioning a distribution range of the first energy consumption volume in the first period into a plurality of first segments by making the first threshold corresponding to at least one preset second threshold which is a boundary value sectioning a distribution range of the second energy consumption volume into a plurality of second segments, respectively. The first threshold calculation means 3 calculates the first threshold so that, with respect to all or a preliminarily selected part of the first segments, a hit ratio which is a ratio of the number of sample values belonging to one corresponding second segment out of the whole number of sample values belonging to one first segment becomes a predetermined value or more.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a heterogeneous energy consumption estimation apparatus for estimating the consumption of a certain energy of a consumer consuming a plurality of types of energy such as city gas and electric power from the consumption of another energy. And a heterogeneous system that estimates the energy consumption of a certain period of a group of general household consumers who receive energy from a power company, etc., from city gas for household use, electricity, etc., from the consumption of another energy The present invention relates to an energy consumption estimation device.

  Conventionally, various methods have been proposed for estimating energy consumption such as electric power and city gas or for predicting energy demand (see, for example, Patent Documents 1 to 4 below). In patent document 1, the estimation method of the consumption per unit time in the arbitrary periods for every consumer is proposed. Patent Document 2 proposes a prediction system for energy demand when energy is supplied to a plurality of energy consumers from an energy factory via an energy supply network. Patent Document 3 proposes a power consumption amount management device that estimates the power consumption amount of each power consuming device model. Patent Document 4 proposes a power usage estimation system for high-load equipment in a night time zone. In any of Patent Documents 1 to 4, the estimation of energy consumption is performed based on the actual data of the same kind of energy, that is, the estimation of power consumption, based on the actual data of power consumption. Yes.

JP 2006-11715 A JP 2007-241360 A JP 2012-90502 A JP 2014-138455 A

  As described above, in the conventional energy consumption estimation or energy demand prediction, the energy type of the estimation or prediction target and the energy type of the actual data used for the estimation or prediction are the same. Conventionally, it has been considered that it is difficult to estimate different types of energy consumption (for example, power consumption from city gas consumption) from a certain energy consumption. This is because the energy consumption patterns in general households differ greatly depending on the energy species.

  Conventionally, there has been almost no request for estimating different types of energy consumption from a certain amount of energy consumption. However, with the liberalization of energy markets such as electric power and city gas, only one energy type supplier such as an electric power company or city gas company can enter into the supply of heterogeneous energy. If it reaches a wide range of consumers in general households, it will be important to estimate the consumption of the different energy.

  In view of the above problems, an object of the present invention is to provide a heterogeneous energy consumption estimation device capable of estimating a different energy consumption from a certain energy consumption with practical accuracy with reduced estimation errors.

In order to achieve the above object, a heterogeneous energy consumption estimation apparatus according to the present invention provides the first period of each consumer in a group of consumers who consume both different first energy and second energy. Based on the known sample data in which the pair of the second energy consumption in the second period that is the same as or different from the first period is an individual sample value, the known energy of the first period A heterogeneous energy consumption estimation device for estimating an unknown consumption of the second energy in the second period from the consumption of the first energy,
Storage means for storing the sample data, and one or more first threshold values which are boundary values for dividing the distribution range of the consumption amount of the first energy in the first period into a plurality of first segments, First threshold value calculating means for calculating each of the second energy consumption distribution ranges corresponding to one or more preset second threshold values which are boundary values for dividing the second energy consumption distribution range into a plurality of second segments. With
The first threshold value calculation means corresponds to the first number corresponding to the whole number of the sample values belonging to one first segment for all of the first segment or a part selected in advance. The first characteristic is that the first threshold value is calculated so that the hit ratio, which is the ratio of the number of sample values belonging to two segments, is equal to or greater than a predetermined value.

  According to the first feature configuration, in all of the first segment or in each of a part selected in advance, the sample values distributed in the first segment correspond to the first ratio corresponding to the hit rate equal to or greater than the predetermined value. When the consumer's second energy consumption of the first energy consumption of the first segment is estimated, the second energy consumption of the corresponding second segment is present. Therefore, the estimation accuracy of the consumption amount of the second energy can be suppressed to a certain error range. That is, the estimation error can be reduced as compared with the case where the distribution ranges of the consumption amounts of the first and second energy are not divided into the first and second segments, respectively.

  In the heterogeneous energy consumption estimation apparatus having the first characteristic configuration, the first threshold value calculation means sets the hit rate to be equal to or greater than the predetermined value for each of the preselected portions of the first segment. And calculating the first threshold value so that the number of hits, which is the number of sample values belonging to both the one second segment corresponding to the one first segment, is maximized within a range of The second feature.

  According to the second characteristic configuration, it is possible to uniquely determine the first threshold value that defines the range of the first segment in which the hit rate is equal to or greater than the predetermined value. In addition, since the number of hits is maximized, it is possible to further suppress the estimation error of the consumption amount of the second energy for the first segment.

  In the heterogeneous energy consumption estimation apparatus having the first or second characteristic configuration, the aggregate is divided into a plurality of groups based on one or more attributes of the consumer, and the storage means includes the individual The sample data is stored so that it can be identified to which of the plurality of groups the sample value belongs, and the one or more second threshold values are set in common through the plurality of groups, and the first threshold value It is a third feature that the calculating means calculates the first threshold value corresponding to the second threshold value for each group.

  According to the third feature configuration, since the first segment is more appropriately defined according to the consumer attribute for each group, it is possible to further suppress the estimation error of the consumption amount of the second energy.

  The heterogeneous energy consumption estimation apparatus having the first or second characteristic configuration receives, as an input, the consumption amount of the first energy of the first consumer selected from the aggregate as the input. A second energy consumption calculating means for calculating a consumption amount of the second energy in the second period using a calculation formula set for every two segments, wherein the second energy consumption calculating means is used as an input; The received amount of the first energy is compared with the first threshold value to determine which one of the plurality of first segments belongs, and the second corresponding to the determined first segment. A fourth feature is that the consumption amount of the second energy in the second period is calculated using the calculation formula set for the segment.

The heterogeneous energy consumption estimation apparatus of the third characteristic configuration includes the consumption amount of the first energy of the one consumer selected from the aggregate and the group to which the consumer belongs. A second energy consumption calculating unit that receives identification information as an input and calculates a consumption amount of the second energy in the second period using a calculation formula set for each second segment;
The second energy consumption amount calculation means compares the first energy consumption amount received as an input with the first threshold value calculated for the group of the identification information received as an input. The second segment in the second period is determined using the calculation formula set for the second segment corresponding to the determined first segment. The fifth feature is to calculate the energy consumption.

  According to the fourth or fifth characteristic configuration, when estimating the second energy consumption corresponding to a certain first energy consumption, the first energy consumption belongs to the first segment to which the first energy consumption belongs. Since the corresponding second segment is determined with a hit rate equal to or greater than a predetermined value and the second energy consumption is estimated using a calculation formula specialized for the second segment, the second energy consumption is estimated. The accuracy can be suppressed within a certain error range. That is, the effects of the first to third feature configurations can be more specifically exhibited.

Furthermore, the heterogeneous energy consumption estimation apparatus according to the present invention provides the known first energy of the first period for each consumer in a set of consumers consuming both different first energy and second energy. A heterogeneous energy consumption estimation device for estimating an unknown consumption amount of the second energy in a second period that is the same as or different from the first period from a consumption amount,
Storage means for storing one or more first threshold values which are boundary values for dividing the distribution range of the consumption amount of the first energy in the first period into a plurality of first segments; and one selected from the aggregate Second energy consumption calculation that accepts the consumption amount of the first energy of the first period of the consumer as an input and calculates the consumption amount of the second energy of the second period using a predetermined calculation formula With means,
A pair of known first energy consumption of the first period and known second energy consumption of the second period for each consumer in the aggregate as individual sample values;
The first threshold corresponds to one or more second thresholds, each of which is a boundary value that divides the distribution range of the second energy consumption amount in the second period into a plurality of second segments, and the first segment The ratio of the number of sample values belonging to one corresponding second segment out of the total number of sample values belonging to one first segment with respect to all or part of a preselected portion A certain hit rate is set to be greater than or equal to a predetermined value,
The calculation formula is set for each second segment,
The second energy consumption calculating means determines which of the plurality of first segments belongs to the first energy consumption received as an input in comparison with the first threshold, A sixth feature is that the consumption amount of the second energy in the second period is calculated using the calculation formula set for the second segment corresponding to the determined first segment.

Furthermore, the heterogeneous energy consumption estimation apparatus according to the present invention provides the known first energy of the first period for each consumer in a set of consumers consuming both different first energy and second energy. A heterogeneous energy consumption estimation device for estimating an unknown consumption amount of the second energy in a second period that is the same as or different from the first period from a consumption amount,
The aggregate is divided into a plurality of groups based on one or more attributes of the consumer;
One or more first threshold values which are boundary values for dividing the distribution range of the consumption amount of the first energy in the first period into a plurality of first segments, and each of the consumers belongs to any of the plurality of groups Storage means for storing information that can be identified, consumption amount of the first energy of the one consumer selected from the aggregate in the first period, and identification information of the group to which the consumer belongs And a second energy consumption calculation means for calculating the consumption amount of the second energy in the second period using a predetermined calculation formula,
A pair of known first energy consumption of the first period and known second energy consumption of the second period for each consumer in the aggregate as individual sample values;
The first threshold corresponds to one or more second thresholds, each of which is a boundary value that divides the distribution range of the second energy consumption amount in the second period into a plurality of second segments, and the first segment The ratio of the number of sample values belonging to one corresponding second segment out of the total number of sample values belonging to one first segment with respect to all or part of a preselected portion It is set for each group so that a certain hit rate is above a predetermined value.
The calculation formula is set for each second segment,
The second energy consumption calculating means compares the first energy consumption received as an input with a plurality of the first threshold values set for the group of the identification information received as an input. The second segment in the second period is determined using the calculation formula set for the second segment corresponding to the determined first segment. The seventh feature is to calculate energy consumption.

  In the heterogeneous energy consumption estimation apparatus of the sixth or seventh feature configuration, the hit ratio is greater than or equal to the predetermined value with respect to each of the first threshold values selected in advance in the first segment. It is set that the number of hits, which is the number of the sample values belonging to both the one second segment corresponding to the one first segment, is maximized in the range of Features.

  According to the sixth to eighth feature configurations, similarly to the fourth or fifth feature configuration, when estimating the consumption amount of the second energy corresponding to the consumption amount of the first energy, A second segment corresponding to the first segment to which the first energy consumption belongs is determined with a hit rate equal to or greater than the predetermined value, and the second energy consumption is calculated using a calculation formula specialized for the second segment. Therefore, the estimation accuracy of the consumption amount of the second energy can be suppressed to a certain error range.

  In the heterogeneous energy consumption estimation apparatus having any one of the fourth to eighth characteristic configurations, the calculation formula uses the second energy consumption in the second period as an objective variable, and the first energy consumption in the first period. A ninth feature is that the regression equation is set for each of the second segments using the consumption amount of one energy as an explanatory variable.

  In the heterogeneous energy consumption estimation device of the ninth feature configuration, the regression equation further includes at least one of one or more attributes of the one consumer selected from the aggregate as an explanatory variable. The tenth feature is that it is a regression equation.

  In the heterogeneous energy consumption estimation apparatus according to the ninth or tenth feature configuration, it is preferable that the regression equation is a regression equation derived by regression analysis using the sample values belonging to the corresponding second segment. Eleven features.

  The heterogeneous energy consumption estimation device having any one of the fourth to eleventh feature configurations selects the consumers one by one from the aggregate, and the first energy of the consumer for the first period is selected. A second energy consumption total that inputs consumption into the second energy consumption calculation means and totals the consumption of the second energy in the second period that is sequentially output from the second energy consumption calculation means. A twelfth feature is provided with the means.

Furthermore, the heterogeneous characteristic estimation apparatus according to the present invention includes a first characteristic value of one of the two events and a second characteristic value of the other of the two events in a set of individuals exhibiting predetermined characteristics for two different events. A heterogeneous characteristic estimation device that estimates an unknown second characteristic value from a known first characteristic value based on known sample data with pairs as individual sample values,
Storage means for storing the sample data; one or more first threshold values which are boundary values for dividing the distribution range of the first characteristic value into a plurality of first segments; and a plurality of distribution ranges of the second characteristic value. First threshold value calculating means for calculating each of the second threshold values corresponding to one or more preset second threshold values, which are boundary values to be divided into second segments,
The first threshold value calculation means corresponds to the first number corresponding to the whole number of the sample values belonging to one first segment for all of the first segment or a part selected in advance. The first characteristic is that the first threshold value is calculated so that the hit ratio, which is the ratio of the number of sample values belonging to two segments, is equal to or greater than a predetermined value.

  According to the first feature configuration, in all of the first segment or in each of a part selected in advance, the sample values distributed in the first segment correspond to the first ratio corresponding to the hit rate equal to or greater than the predetermined value. Probability of existing within the range of the second characteristic value of the corresponding second segment when estimating the second characteristic value of the individual of the first characteristic value of the first segment. Therefore, the estimation accuracy of the second characteristic value can be suppressed within a certain error range. That is, the estimation error can be reduced as compared with the case where the distribution ranges of the first and second characteristic values are not divided into the first and second segments, respectively.

  In the heterogeneous property estimation apparatus having the first characteristic configuration, the first threshold value calculation unit has a range in which the hit ratio is equal to or more than the predetermined value for each of the preselected portions of the first segment. And calculating the first threshold value so that the number of the sample values belonging to both the first segment and the second segment corresponding to the first segment is maximized. It is characterized by.

  According to the second characteristic configuration, it is possible to uniquely determine the first threshold value that defines the range of the first segment in which the hit rate is equal to or greater than the predetermined value. In addition, since the number of hits is maximized, the estimation error of the second characteristic value for the first segment can be further suppressed.

  In the heterogeneous property estimation apparatus having the second or third feature configuration, the aggregate is divided into a plurality of groups based on one or more attributes of the individual, and the storage means includes the individual sample values. The sample data is stored so that it can be identified to which of the plurality of groups, one or more second threshold values are set in common through the plurality of groups, and the first threshold value calculating means The third feature is that the first threshold value corresponding to the second threshold value is calculated for each group.

  According to the third feature, the first segment is more appropriately defined according to the attribute of the individual for each group, so that the estimation error of the second characteristic value can be further suppressed.

  According to the heterogeneous energy consumption estimation apparatus having the above characteristic configuration, it is possible to estimate the heterogeneous energy consumption from a certain energy consumption with practical accuracy with suppressed estimation errors.

The block diagram which shows schematic structure of one Embodiment of the heterogeneous energy consumption estimation apparatus which concerns on this invention Scatter plots with normalized values of monthly city gas consumption and power consumption for individual consumers as sample points The flowchart which shows the rough process sequence of the 1st process by a 1st threshold value calculation means. The figure which illustrates typically the processing content of step # 12 of the 1st process shown in FIG. The figure which shows an example of the process result of 1st process typically The flowchart which shows the rough process sequence of the 2nd process by an electric power consumption calculation means. Table showing the results of evaluating the estimation error by the apparatus of the present invention according to the presence / absence of the first processing, the presence / absence of group classification, and the difference in calculation formula Table showing error improvement effect due to difference in first processing, group classification, and calculation formula for estimation error shown in FIG.

  Hereinafter, an embodiment of a heterogeneous energy consumption estimation apparatus (hereinafter referred to as “the present invention apparatus” as appropriate) according to the present invention will be described with reference to the drawings.

  The device of the present invention is intended for a group of consumers (population) consuming a plurality of types of energy such as city gas and electric power, and the first energy (for example, the plurality of types of energy in a certain first period) , City gas), the consumption amount of the second energy (for example, electric power) of the plurality of types of energy in the second period that is the same as or different from the first period is estimated. In the following explanation, city gas is assumed as the first energy, electric power is assumed as the second energy, one month is assumed as the first period and the second period, and the first period and the second period are the same month. Is assumed. That is, for example, when the first period is X month (X = 1 to 12), the second period is also X month. Further, it is assumed that a general household that receives a supply of city gas and power for a general household is assumed as a consumer consuming a plurality of types of energy.

  FIG. 1 is a block diagram showing an example of the configuration of the device 1 of the present invention. As shown in FIG. 1, the device 1 of the present invention includes a storage unit 2, a first threshold value calculation unit 3, a power consumption calculation unit 4, a power consumption totaling unit 5, and a calculation formula deriving unit 6. ing. The device 1 of the present invention is configured by using, for example, hardware resources (CPU, storage device, interface device, etc.) and software resources (OS, programs, etc.) of a general-purpose computer such as a personal computer. 3. The power consumption calculation means 4, the power consumption totalization means 5, and the calculation formula derivation means 6 are provided in the computer, and are executed by executing a program for processing each part separately by the computer. To be realized. The storage means 2 may be configured using a nonvolatile storage device (hard disk drive (HDD), solid state drive (SSD), etc.) built in a general-purpose computer, but the data to be stored is shared with other systems. In this case, a configuration may be adopted in which a database is provided outside the general-purpose computer and connected via a data communication path such as a local area network (LAN). The power consumption calculating means 4 and the power consumption calculating means 5 correspond to the second energy consumption calculating means and the second energy consumption calculating means.

  In the present embodiment, the data stored in the storage device 1 includes the consumer identification code, the actual or estimated value of the city gas consumption for each month, and the power consumption for each month for each consumer of the aggregate. It consists of an actual value or an estimated value, and a group number which is group identification information indicating the attribute classification of the consumer. Here, a pair of each actual value of the city gas consumption and the power consumption of each month for each consumer is referred to as a sample value, and a set of the sample values of the entire aggregate is referred to as sample data.

  In the present embodiment, each consumer in the collection is classified into a plurality of groups according to some attributes of the consumer. As the attribute, for example, a residence type (a detached house / apartment), a family structure, the presence / absence of a specific power or gas consuming device, the presence / absence of a distributed power generation device (solar panel or the like), and the like are assumed. The number of attributes to be used and the total number of groups are arbitrary, but if the number of groups is too large relative to the total number of consumers, the number of samples in one group becomes small, so the number of samples in each group becomes too small. Adjust so that there is no. In addition, since the appropriate range of the number of samples of each group is related also to the number of segments which will be described later, it is preferable to set appropriately for the characteristics of the aggregate to be used.

  FIG. 2 shows a city gas that is a sample value of each consumer in a two-dimensional coordinate space in which the vertical axis is normalized power consumption of a certain month and the horizontal axis is normalized city gas consumption of the same month. The scatter diagram which plotted the value which normalized consumption and electric power consumption as a sample point is shown. In the example shown in FIG. 2, the number of sample points is about 8000. Normalized power consumption and normalized city gas consumption are the values obtained by dividing the power consumption and city gas consumption of each sample value by the maximum values of power consumption and city gas consumption of each sample value, respectively. It is. Hereinafter, the normalized power consumption and the normalized city gas consumption are referred to as a power consumption E and a city gas consumption G, respectively, as appropriate. In the example shown in FIG. 2, the power consumption E and the city gas consumption G are such that the power consumption E is in the range of about 0.02 to 0.4 and the city gas consumption G is in the range of about 0 to 0.45. This clearly shows a positive correlation.

  Next, the contents of the process (first process) of the first threshold value calculation means 3 will be described with reference to FIGS. FIG. 3 is a flowchart showing a processing procedure of the first processing. FIG. 4 is a diagram schematically illustrating the processing content of step # 12 shown in FIG.

  In the present embodiment, the power consumption to be estimated is divided in advance into a plurality of segments (second segments) before executing the first process. The boundary values of each second segment (second threshold TEi, i = 1 to m−1, where m is the number of segments so that the sample points showing the positive correlation described above are included in each divided second segment. ) Is set in advance. The second threshold value TE0 = 0 and the second threshold value TEm = 1 are set to i = 0 to m. In the following description, for the sake of convenience, assuming that power consumption is divided into four second segments, each second segment has a power consumption E of 0 to TE1, TE1 to TE2, TE2 to TE3, TE3. There are four ways of ~ 1. In the example shown in FIG. 2, the second threshold value TE3 is set to around 0.2 in order to include the above-described sample points showing the positive correlation in the second segment (TE3 to TE1). become. TE1 and TE2 are set on the same basis. In the present embodiment, the second threshold values TE1 to TE3 are common among the groups.

  The 1st process by the 1st threshold value calculation means 3 is a process which divides | segments city gas consumption into the same number of segments (1st segment) as electric power consumption. When the number of segments m is 4, each first segment has four types of city gas consumption G of 0 to TG1j, TG1j to TG2j, TG2j to TG3j, and TG3j to 1 for each group. j is an ordinal number indicating the group, and is 1 to n (n is the total number of groups). In the first process, first threshold values TG1j to TG (m−1) j, which are boundary values of the first segments, are associated with the second threshold values TE1 to TE (m−1), respectively. This is a process for determining each group so as to satisfy the second classification criterion.

  Hereinafter, the first process will be described along the flow shown in FIG. In the following description, the order numbers of the first segment and the second segment are 1, 2,... M in the order in which the city gas consumption G and the power consumption E increase from 0 to 1.

  First, in step # 10, variables i and j are initially set to (m-1) and 1. Subsequently, in step # 11, the first threshold value TGij is provisionally set to the initial value TG0ij. For example, TG0ij is 1 when i = m−1, and is 1st threshold TG (i + 1) j determined in the loop one time before variable i when i = 1 to m−2. .

  Subsequently, in step # 12, while reducing the first threshold value TGij step by step in a predetermined minute step, the hit rate Rij shown in the following equation 1 is calculated.

  (Expression 1) Rij = Nij / N0ij × 100 (%)

  N0ij in Equation 1 is a city gas consumption amount G within the range of TGij to TG0ij (corresponding to the region obtained by adding region A and region B in FIG. 4), that is, within the provisional (i + 1) th first segment In the total number of samples in Nij, the power consumption E is within the range of TEi to TE (i + 1) (in the (i + 1) th second segment, corresponding to the region A in FIG. 4). It is a certain number of samples. Hereinafter, Nij is referred to as “number of hits”. TE (i + 1) is 1 when i = m−1.

  While the target hit rate Rij is equal to or higher than a predetermined reference value (for example, 70%) and maintains the reference value or higher, N0ij and Nij are counted while decreasing the first threshold value TGij stepwise. The rate Rij is calculated. Normally, city gas consumption G and power consumption E have a positive correlation. Therefore, if the first threshold value TGij is decreased and the range of TGij to TG0ij is expanded, N0ij and Nij increase, but N0ij increases. Since the degree of increase is large, the hit rate Rij decreases. Then, TGij that satisfies the first classification criterion that the hit rate Rij is equal to or greater than the reference value and that satisfies the second classification criterion that the target number Nij is the maximum is determined as the first threshold value TGij. Various algorithms are conceivable as an algorithm for finding TGij that satisfies the first and second classification criteria, and since the algorithm is not the gist of the present invention, a detailed description is omitted.

  The purpose of applying the second classification criterion is to estimate the power consumption E estimation error by maximizing the hit number Nij when the first threshold value TGij that satisfies the first classification criterion is not uniquely determined. The first threshold value TGij is uniquely determined while suppressing the above. However, if the first threshold value TGij that satisfies the first classification criterion is not determined in the (i + 1) th first segment, the second classification criterion is ignored in the previous (i + 2) th first segment. In some cases, the already determined first threshold value TG (i + 1) j is increased stepwise to adjust the first threshold value TGij that satisfies the first classification criterion in the (i + 1) th first segment. obtain.

  If it is determined in step # 13 that i ≠ 1, i is decreased by 1 in step # 14, and steps # 11 to # 14 are repeated until the variable i is determined to be 1. All threshold values TGij (i = 1 to m−1) are derived.

  If it is determined in step # 13 that i = 1, it is determined in step # 15 whether variable j is n. If variable j is not n, variable j is incremented by 1 in step # 16. i is initialized to (m−1), and steps # 11 to # 14 are repeated to derive the first threshold value TGij (i = 1 to m−1) of the next group j. Thereafter, steps # 11 to # 16 are repeated until the variable j is determined to be n in step # 15, and the first threshold TGij (i = 1 to 1) is set for all groups j (j = 1 to n). m-1) is derived.

  FIG. 5 schematically shows a processing result in a certain group when the number of segments m is 4, as an example. By executing the first process of the first threshold value calculation means 3, the relationship between the first segment of city gas consumption and the second segment of power consumption is determined for each group. If the city gas consumption amount is known, it is possible to identify with a certain probability which second segment the consumer's power consumption corresponds to. Also, by increasing the number of segments m, the resolution of the first segment with i = 2 to m−2 is increased, so that the accuracy of estimating the power consumption is improved.

  Next, the contents of the process (second process) of the power consumption calculation means 4 will be described with reference to FIG. FIG. 6 is a flowchart showing a processing procedure of the second processing.

  In step # 20, the city gas consumption of one consumer is received together with the group number, and first, the group to which the consumer belongs is specified. Furthermore, the city gas consumption G is calculated by normalizing the received city gas consumption. In addition, when accepting the city gas consumption G instead of the city gas consumption, the normalization is not necessary. The city gas consumption is stored in the storage device 1 or another storage device in advance, and is read from the storage device 1 or another storage device by the control from the outside or the power consumption totaling means 5, It may be configured to input to the consumption calculation means 4 or may be configured to input directly to the power consumption calculation means 4 by an input operation of an operator or the like.

  Next, in step # 21, it is specified to which of the first segments determined for the group specified in the first process the city gas consumption amount G belongs.

  Next, in step # 22, the calculation formula of the second segment corresponding to the first segment specified in step # 21 is selected from the preset calculation formulas for each second segment, and calculated in step # 20. The power consumption E is calculated by substituting the received city gas consumption G into the selected calculation formula.

  The processes of steps # 20 to # 22 are executed for all consumers to be estimated. The power consumption totalization means 5 aggregates the power consumption E calculated by the power consumption calculation means 4 for each consumer, and multiplies the maximum power consumption by the calculation target of all consumers to be estimated. Calculate the total power consumption of the month. When this process is performed from January to December, the total power consumption for the year is calculated.

  Next, the calculation formula used in step # 22 will be described. In the present embodiment, the calculation formula derived in advance for each second segment by the calculation formula deriving means 6 is used. As the calculation formula, the following three types of calculation formulas are assumed to be used. The first calculation formula is a linear expression shown in the following formula 2, the second calculation formula is a simple regression formula shown in the following formula 3, and the third calculation formula is expressed by the following formula 4. It is the multiple regression equation shown.

(Equation 2) E = a1 × G + a0
(Equation 3) E = b1 × G + b0
(Equation 4) E = c1 × G + c2 × F2 +... + Ck × Fk + c0

  In Eqs. 2 to 4, E is power consumption E, and G is city gas consumption G. In the single regression equation of Equation 3, the power consumption E is an objective variable, the city gas consumption G is an explanatory variable, and the number of explanatory variables is 1. In the multiple regression equation of Equation 4, the power consumption E is an objective variable, the city gas consumption G is an explanatory variable, and in addition to the city gas consumption G, it relates to (k-1) consumer attributes. The explanatory variables F2 to Fk are used, and the number of explanatory variables is k (k is an integer of 2 or more). The explanatory variables F2 to Fk may be all or a part of the attributes used for the group classification described above, and may further add attributes that are not used for the group classification. Therefore, the explanatory variables F2 to Fk may be discrete variables. a1, b1, and c1 are coefficients of the city gas consumption G, and a0, b0, and c0 are constants. c2 to ck are coefficients of the explanatory variables F2 to Fk.

  Hereinafter, the derivation method of each calculation formula shown in Formula 2 to Formula 4 by the calculation formula deriving means 6 will be described. When the calculation formula is a linear formula shown in Formula 2, the calculation formula is derived for each second segment and each group. Specifically, the first threshold values TG1j to TG (m−1) j corresponding to the second threshold values TE1 to TE (m−1) determined for each group in the first process described above are set as the second segment. Each time, that is, two second threshold values TEi and second threshold value TE (i + 1) that define the second segment, and two second threshold values TGij and first threshold value TG (i + 1) that define the corresponding first segment. Substituting j into the first equation, the coefficient a1 and the constant a0 are calculated for each second segment and each group. In other words, the calculation formula shown in Formula 2 is represented by a broken line shown by a broken line in FIG. In addition, in the case of using the linear expression shown in Equation 2 as the calculation formula in Step # 22 of the second process, it corresponds to the group specified in Step # 20 and the first segment specified in Step # 21. The calculation formula for the second segment is selected and used.

  In the case where the calculation formula is the single regression formula shown in Equation 3, the calculation formula is derived by performing a single regression analysis on all the groups for each second segment. Specifically, for the i-th second segment defined by the second threshold value TE (i−1) and the second threshold value TEi, all the sample values existing in the i-th second segment are included. Using the power consumption E and the city gas consumption G, for example, a well-known single regression analysis by the least square error method is performed. If the number of samples existing in the i-th second segment is also sufficient for the regression analysis, the i-th first number is also present in the corresponding i-th first segment. The single regression analysis may be performed using the power consumption E and the city gas consumption G of sample values that exist in two segments and also exist in the corresponding i-th first segment. Various methods other than the above can be considered as the method of selecting the sample value for securing the number of samples necessary for the single regression analysis.

  When the calculation formula is the multiple regression equation shown in Formula 4, the calculation formula is derived by performing multiple regression analysis for all the groups for each second segment. Specifically, for the i-th second segment defined by the second threshold value TE (i−1) and the second threshold value TEi, all the sample values existing in the i-th second segment are included. A well-known multiple regression analysis by the least square error method, for example, is performed using the power consumption E, the city gas consumption G, and the values of the explanatory variables F2 to Fk relating to the consumer attributes of the sample value. If the number of samples existing in the i-th second segment is also sufficient for the regression analysis, the i-th first number is also present in the corresponding i-th first segment. The multiple regression analysis may be performed using the power consumption E and the city gas consumption G of sample values that exist in two segments and also exist in the corresponding i-th first segment. Various methods other than the above can be considered as a method of selecting sample values for securing the number of samples necessary for the multiple regression analysis. When the multiple regression equation shown in Equation 4 is used as the calculation formula in step # 22 of the second process, when the city gas consumption of one consumer is received together with its group number in step # 20. The values of the explanatory variables F2 to Fk relating to the consumer attributes are also received at the same time, and in step # 22, the explanatory variables F2 to Fk are calculated in step # 20 or selected together with the received city gas consumption G. By substituting into the equation, the power consumption E is calculated. When the group number and the values of the explanatory variables F2 to Fk are uniquely associated, a table indicating the association is prepared separately, and the values of the explanatory variables F2 to Fk are determined from the group numbers received by the table. May be read out.

  Among the calculation formulas shown in the above formulas 2 to 4, the multiple regression formula shown in the formula 4 is the best with respect to the estimation accuracy of the power consumption E, and the primary formula shown in the formula 2 and the formula 3 are shown. There is no significant difference in the single regression equation. However, when the positive correlation between the power consumption E and the city gas consumption G of the sample value to be used is strong, there is a possibility that high estimation accuracy can be obtained even with the calculation formula shown in Equation 2 or Equation 3.

  Next, with respect to the estimation accuracy by the apparatus 1 of the present invention, the results of comparison and examination of the difference due to the presence / absence of the first process, the difference due to the presence / absence of group classification, and the difference due to the difference of the calculation formulas shown in Equations 2 to 4 are A description will be given with reference to FIG.

  When the first process is not performed, the power consumption E and the city gas consumption G are not divided into segments. Therefore, the calculation formula used in the second process is the single regression equation shown in Equation 3 or Equation 4. The multiple regression equation is shown. When performing the first process, the presence or absence of group classification is distinguished. As for the presence / absence of group classification, when the group classification is performed, the first threshold value TGij (i = 1 to m−1) is derived for each group j because the first process is performed for each group. Also in the process, identification of the first segment in step # 21 and selection of the calculation formula when using the calculation formula shown in Formula 2 are performed according to the group j. On the other hand, when the group is not classified, the first process is not executed for each group, so the first threshold value TGij (i = 1 to m−1) is derived in one way regardless of the group, Also in the second process, identification of the first segment in step # 21 and selection of the calculation formula when using the calculation formula shown in Equation 2 are performed regardless of the group.

The comparative study was performed for the following eight cases A to H. Cases A and B are comparative examples for the device 1 of the present invention, and cases C to H are examples of the device 1 of the present invention.
Case A: first process = no, calculation formula = single regression equation shown in Formula 3Case B: first process = No, calculation formula = multiple regression formula shown in Formula 4Case C: first process = Yes, Group classification = None, Calculation formula = Primary expression shown in Formula 2Case D: 1st process = Yes, Group classification = No, Calculation formula = Single regression formula shown in Formula 3Case E: 1st process = Yes, Group classification = None, Calculation formula = Multiple regression equation shown in Formula 4Case F: First processing = Yes, Group classification = Yes, Calculation formula = Primary formula shown in Formula 2Case G: First processing = Yes, Single regression equation shown in group classification = Yes, calculation formula = Equation 3. Case H: First processing = Yes, group classification = Yes, calculation equation = Multiple regression equation shown in Equation 4.

  For cases A to H, the total power consumption (estimated value) by month from April to December of the given year and from January to March of the following year and the total year The total power consumption (estimated value) was calculated, and the error of each estimated value was calculated from the true value of each total power consumption and the estimated value by the following equation (5). Each calculated error (rounded off after the decimal point) is listed in FIG. The number of samples used in this comparative study was 8000, the number of segments in cases C to H was 5, the number of groups in cases F to H was 20, and the number of explanatory variables in the multiple regression equations for cases B, E, and H (city) The gas consumption G is 4). Further, the hit rate Rij ≧ 70% was adopted as the first classification criterion of the first process.

  (Equation 5) Error = | true value−estimated value | / true value × 100 (%)

  Further, the error improvement effect of performing the first processing (case A to case D, case B to case E), the error improvement effect of performing group classification (case C to case F, case D to case G, case E) To case H), and the error improvement effect (case C to case D, case F to case) of changing the linear equation of equation 2 to the simple regression equation of equation 3 (hereinafter referred to as “single regression equation” as appropriate) G), the error improvement effect of changing the single regression equation of Equation 3 to the multiple regression equation of Equation 4 (hereinafter referred to as “multiple regression equation” as appropriate) (Case A to Case B, Case D to Case E, Case G to Case H), error improvement effect of performing the first process and group classification (case A to case G, case B to case H), error improvement effect of performing the first process and multiple regression formula ( Case A to Case E), group Figure 2 shows the error improvement effect of performing the multiple regression formulation (case D to case H), and the error improvement effect of the first process, group classification and multiple regression formulation (case A to case H). 8 is displayed as a list. In addition, the error improvement effect was evaluated by the error improvement rate (rounded to the nearest decimal place) shown in Equation 6 below. In Equation 6, error X is an error in the case before improvement, and error Y is an error in the case after improvement. For example, in the case of the error improvement effect (case A to case H) of performing the first process, the group classification, and the multiple regression formula, the error X is the error of case A and the error Y is the error of case H.

  (Equation 6) Error improvement rate = (error X−error Y) / error X × 100 (%)

  As shown in FIG. 7, in case A (comparative example) in which a single regression equation is used without performing the first process and group classification, the monthly error is 29% to 36%, and the monthly error average is 32%. The annual error is 30%, and the error exceeds 30% with a high probability. As an error in estimating such total monthly power consumption or annual total power consumption, an error exceeding 30% is not preferable. However, in case D, in which the first process is added to case A, the monthly error has decreased to 25% to 32%, the monthly error has averaged 28%, and the annual error has decreased to 26%. The average of the error improvement rate is 12%. On the other hand, in case B (comparative example) in which the first process is not performed for case A, the monthly error is 25% to 31%, the average of the monthly error is 28%, The error decreases to 26%, and the average error improvement rate by month is 11%. Here, it can be seen that the error improvement effect by adding the first process to Case A is almost the same as the error improvement effect when the multiple regression equation is formed without performing the first process, but is slightly larger. In both cases B and D, the monthly error exceeds 30% in the cold season when city gas consumption is high from January to March, but overall, the error is improved by 11% or more. Furthermore, in case E, in which the first process and multiple regression formula are performed on case A, the monthly error is 21% to 29%, the monthly error is 25% average, and the annual error is 22%. The monthly error is less than 30% throughout the year from January to December. As described above, from the results of Case D and Case E, the error improvement rate due to the addition of the first process is 11% or 12% regardless of whether the calculation formula is a single regression formula or a multiple regression formula. By performing both of the multiple regression equations, the error improvement rate becomes 20%, and further error improvement effect can be expected.

  Next, the error improvement effect by performing group classification will be examined. In case C, which is a linear expression in which the first processing is added to case A and the unclassified calculation formula is shown in Equation 2, the monthly error is 26% to 32%, the monthly error average is 29%, Whereas the annual error is 26%, in case F with group classification added to case C, the monthly error is 24% to 29%, the monthly error is 26% average, and the annual error is Each of them decreased to 23%, and there was a clear error improvement effect from case C to case F, and the average of the monthly error improvement rates was 10%. In addition, in the case D of the single regression equation shown in Formula 3 in which the first process is added to the case A and the calculation formula without group classification is Equation 3, the monthly error is 25% to 32%, and the monthly error average is 28. %, The yearly error is 26%, but in case G, in which group classification is added to case D, the monthly error is 23% to 28%, the average monthly error is 25%, The errors have decreased to 22%, respectively, and a clear error improvement effect from case D to case G is seen, and the average of the monthly error improvement rates is 10%. Further, in case E of the multiple regression equation shown in Formula 4 in which the first process is added to case A and the multiple regression equation shown in Equation 4, the monthly error is 21% to 29%, and the monthly error average is 25. %, The annual error is 22%, but in case H where group classification is added to case E, the monthly error is 20% to 25%, the monthly error is 22%, The errors are reduced to 18%, respectively, and a clear error improvement effect from case E to case H is seen, and the average of the monthly error improvement rates is 13%. From the above, even if the calculation formula is any one of the linear formula shown in Formula 2, the single regression formula shown in Formula 3, and the multiple regression formula shown in Formula 4, the error improvement effect by group classification is clear. In particular, when the calculation formula is the multiple regression formula shown in Formula 4, a more remarkable error improvement effect is seen.

  Next, the error improvement effect by performing both the first processing and the group classification will be examined. As described above, the first process and the group classification have an error improvement effect by themselves, but by performing both at the same time, the error improvement effect of both is added and exhibited. When the error improvement rate by the first process is A (%), the error improvement rate by the group classification is B (%), and the error improvement ratio by both the first process and the group classification is C (%), the error improvement ratio C is And expressed by the following formula 7. However, since the group classification is based on the first process, the error improvement rate by the group classification is based on the error when the first process is performed.

  (Expression 7) C = A + B−A × B / 100 (%)

  From the error improvement rate shown in FIG. 8, the average of the monthly error improvement rate by the first processing from case A to case D is 12%, and the monthly error improvement rate by the first processing from case B to case E is The average is 11%. Further, the average of the monthly error improvement rate by the group classification from case D to case G is 10%, and the average of the monthly error improvement ratio by the group classification from case E to case H is 13%. On the other hand, the monthly error improvement rate by the first processing from the case A to the case G and the group classification is 20%, and the monthly error improvement by the first processing from the case B to the case H and the group classification is 20%. The average of the rate is 22%, and it can be seen that the relationship expressed by Equation 7 is satisfied.

  Next, a case E in which a group regression is formed for the case D without group classification and a case G in which the group classification is added to the case D without changing the calculation formula are compared. In case E, the monthly error is 21% to 29%, the average monthly error is 25%, and the annual error is 22%, whereas in case G, the monthly error is 23% to 28%, The average monthly error is 25%, and the annual error is 22%. The errors in both cases are approximate. The average monthly error improvement rate from case D to case E is 10%, while the average monthly error improvement rate from case D to case G is 10%. It is equivalent. Furthermore, the average of the monthly error improvement rate by performing group classification from case D to case H and multiple regression formulation at the same time is 22%.

  As described above, in addition to the first processing, there is an error improvement effect by making a multiple regression equation without performing group classification, and an error improvement effect by performing group classification with the calculation equation being the single regression equation shown in Equation 3. The error improvement effect obtained by performing group classification in addition to the first process, and further forming a multiple regression formula is more than the sum of the above two error improvement effects. I understand that.

  Next, the error improvement effect by making a single regression equation will be examined. In the case where there is no group classification, in the case C of the primary expression shown in Equation 2, the monthly error is 26% to 32%, the monthly error average is 29%, and the annual error is 26%. In case D of the simple regression equation shown in Equation 3, the monthly error is 25% to 32%, the average monthly error is 28%, and the annual error is 26%. The average of the monthly error improvement rate based on the single regression formula is only 2%, and the error improvement rate is negative in the error for May and the year, and the error is smaller in Case C. In Case C and Case D, there is no significant difference in estimation accuracy. In the case of the group classification, in the case of the primary expression shown in Equation 2, the monthly error is 24% to 29%, the monthly error average is 26%, and the annual error is 23%. In case G of the single regression equation shown in Equation 3, the monthly error is 23% to 28%, the average monthly error is 25%, and the annual error is 22%. The average error improvement rate by month based on the single regression formula is only 2%, and the error improvement rate (evaluation after the decimal point) is negative for the errors in February, April, July and October. Case F has a smaller error and Case F and Case G show no significant difference in estimation accuracy.

  As described above, the error improvement effect by the first process, the error improvement effect by the group classification, and the error improvement effect by the multiple regression formula were confirmed.

  Hereinafter, another embodiment will be described.

  <1> In the above embodiment, the device 1 of the present invention includes the storage unit 2, the first threshold value calculation unit 3, the power consumption calculation unit 4, the power consumption totalization unit 5, and the calculation formula deriving unit 6. In the case where the device 1 of the present invention is used for associating the segment between the power consumption and the city gas consumption, the device 1 of the present invention does not necessarily include the first threshold value calculation means 3, the power The consumption calculation means 4, the power consumption totalization means 5, and the calculation formula deriving means 6 may not be provided.

  Further, when the power consumption for each consumer calculated by the power consumption calculation unit 4 is separately tabulated, the power consumption tabulation unit 5 may be provided separately from the device 1 of the present invention. For example, it is assumed that the power consumption by consumer calculated by the power consumption calculation unit 4 is stored in the storage unit 2.

  Further, in the case of using the power consumption calculation means 4 in a fixed manner without updating the calculation formula, the calculation formula deriving means 6 is provided separately from the apparatus 1 of the present invention, and the calculation formula derived in advance is provided. Is also preferably set in the power consumption calculation means 4.

  Furthermore, the present invention device 1 may be configured not to include the first threshold value calculation means 3. However, in this case, the first threshold value TGij preset by the first threshold value calculation unit 3 is stored in a storage unit that can be used by the storage unit 2 or the power consumption calculation unit 4 and used in the second process. It is preferable to make it possible.

  <2> In the description of the first process of the above embodiment, the power consumption and the city gas consumption are the normalized power consumption E and the city gas consumption G, respectively. Actual consumption amounts may be processed.

  In the description of the first process of the above embodiment, the first and second classification criteria are satisfied in Step # 12 for the first first segment whose city gas consumption G is 0 to TG1j. It is not determined whether or not the first classification criterion is satisfied in the first first segment. Accordingly, in the first first segment, a step for determining whether or not the first classification criterion is satisfied is separately provided. In the determination, when the first classification criterion is not satisfied, the second classification is performed. In step # 12 for the first segment, on the condition that the first and second first segments simultaneously satisfy the first classification criterion, so as to satisfy the second classification criterion for the second first segment. The first threshold TG1j may be set (treatment 1). Further, in this case, if the first threshold value TG1j that satisfies the first reference at the same time for the first and second first segments cannot be set at the same time, the second threshold value TE1 set in the preprocessing of the first process is set. Then, the first process may be executed again (step 2).

  Further, when performing the above-described treatment 1 and treatment 2, since the target number N1j in the second first segment decreases from the initial maximum value, the decrease is assigned to the first segment from the third to the m-th. The first threshold value TG2j to the first threshold value TG (m−1) j may be reset so that only the target number N1j in the second first segment is not extremely decreased. Therefore, in the first first segment, when the first classification criterion is not satisfied, the treatment that satisfies the second classification criterion in Step # 12 is a provisional treatment.

  Furthermore, in the description of the first process of the above embodiment, the first threshold value TGij is set in order from the mth first segment to the second first segment, but conversely, the first first segment To the (m−1) th first segment. However, in step # 12, the first threshold value TGij is increased step by step in a predetermined minute step. Furthermore, in this case, since the possibility that the first classification criterion is not satisfied in the first first segment described above occurs in the m-th first segment, the same treatment as the treatment 1 or the treatment 2 can be performed. Conceivable. Furthermore, regardless of the setting order of the first threshold value TGij, at least one of the first and mth first segments is set as a dummy first segment that allows the first classification criterion not to be satisfied, and the dummy first segment The dummy second segment corresponding to may be set in the pre-process of the first process.

  Furthermore, in the description of the first process of the above embodiment, it is assumed that the first threshold value TGij is set for each group j, but there is a stronger positive correlation between the power consumption E and the city gas consumption G. In some cases or when a multiple regression equation having a higher error improvement effect can be used as a calculation formula, the above-described group classification is not necessarily required. However, since further error improvement effects can be reliably expected by group classification, it is preferable to perform group classification as much as possible.

  <3> In the above embodiment, the first period and the second period are assumed to be one month, and the first period and the second period are assumed to be the same month. If there is data on the combination of the first energy consumption of the first period and the second energy consumption of the second period, the first period and the second period are limited to one month. In addition, it may be a short period (for example, one week, one day, one hour, etc.), and the first period and the second period may not be the same period. For example, when the correlation between the consumption amount of the first energy and the consumption amount of the second energy in the second period is weak in a certain period of one year, it is different from the consumption amount of the second energy in the certain period. When the correlation of the consumption amount of the first energy in the period is stronger, it is preferable to associate the consumption amount of the second energy in the specific period with the consumption amount of the first energy in the other period.

  Furthermore, as the first and second energy handled by the device 1 of the present invention, the case where city gas and power are assumed and the power consumption is estimated from the city gas consumption has been described. It can also be applied to estimate city gas consumption from electricity consumption. Furthermore, one or both of the first and second energy may be an energy species other than city gas and electric power. In this case, there is not necessarily a positive correlation between the first and second energy, and a negative correlation may be used. That is, it is necessary that there is a positive or negative correlation between the first and second energy.

  <4> In the above embodiment, the estimation accuracy by the device 1 of the present invention has been described with reference to FIGS. 7 and 8, but the number of samples, the number of segments, the number of groups, and the explanatory variables of the multiple regression equation used in the description are The number, the lower limit value of the hit rate of the first classification standard, and the like are examples, and the device 1 of the present invention is not limited to the specific example.

  <5> In the above embodiment, the device 1 of the present invention is assumed to be used as a heterogeneous energy consumption estimation device. In the case where there is a positive or negative correlation between the distribution of one first characteristic value and the distribution of the other second characteristic value in a collection of individuals each exhibiting a predetermined characteristic, the one first characteristic value to the other It is also possible to use as a heterogeneous characteristic estimation device for estimating the second characteristic value of That is, the device 1 of the present invention as the heterogeneous energy consumption estimation apparatus of the above embodiment can be regarded as one aspect of the heterogeneous characteristic estimation apparatus.

  In this case, the first processing by the first threshold value calculation unit 3 is performed by setting one or more first threshold values, which are boundary values for dividing the distribution range of the first characteristic value into a plurality of first segments, as the distribution range of the second characteristic value. Is a process of calculating each corresponding to one or more preset second threshold values, which are boundary values that divide into a plurality of second segments. Here, as the first and second classification criteria, those described in the first process of the above embodiment can be used as they are. Further, group classification can also be handled in the same manner as in the above embodiment by performing based on one or more attributes of the individual, and the first threshold value is also set for each group. As an example, one of the first and second characteristic values is energy consumption such as the above-mentioned city gas or electric power, and the other is consumption other than energy, for example, specific household expenses such as food expenses. The

  The heterogeneous energy consumption estimation apparatus according to the present invention can be used to estimate the consumption of another energy from the consumption of a certain energy of a consumer who consumes multiple types of energy such as electric power.

1: heterogeneous energy consumption estimation device 2: storage means 3: first threshold value calculation means 4: power consumption calculation means 5: power consumption aggregation means 6: calculation formula derivation means

Claims (15)

The consumption of the first energy in the first period for each consumer in the aggregate of consumers consuming both the first energy and the second energy different from each other and in the second period that is the same as or different from the first period Based on the known sample data in which the pair of consumption amounts of the second energy is an individual sample value, the unknown second amount of the second period is calculated from the known consumption amount of the first energy of the first period. A heterogeneous energy consumption estimation device for estimating energy consumption,
Storage means for storing the sample data;
One or more first threshold values, which are boundary values that divide the distribution range of the consumption amount of the first energy in the first period into a plurality of first segments, and the distribution of the consumption amount of the second energy in the second period. First threshold value calculation means for calculating each of the threshold values corresponding to one or more preset second threshold values, which are boundary values for dividing the range into a plurality of second segments,
The first threshold value calculation means corresponds to the first number corresponding to the whole number of the sample values belonging to one first segment for all of the first segment or a part selected in advance. The heterogeneous energy consumption estimation apparatus, wherein the first threshold value is calculated so that a hit ratio, which is a ratio of the number of sample values belonging to two segments, is equal to or greater than a predetermined value.   The first threshold value calculation means corresponds to one first segment within a range in which the hit rate is equal to or greater than the predetermined value for each of the preselected portions of the first segment. 2. The heterogeneous energy consumption estimation apparatus according to claim 1, wherein each of the first threshold values is calculated so that a target number that is the number of the sample values belonging to both of the second segments is maximized. . The aggregate is divided into a plurality of groups based on one or more attributes of the consumer;
The storage means stores the sample data so that the individual sample values can be identified to which of the plurality of groups,
The second threshold value of 1 or more is set in common through a plurality of the groups,
3. The heterogeneous energy consumption estimation apparatus according to claim 1, wherein the first threshold value calculation unit calculates the first threshold value corresponding to the second threshold value for each group. The consumption of the first energy of the first period of the one consumer selected from the aggregate is accepted as an input, and using the calculation formula set for each of the second segments, A second energy consumption calculating means for calculating the consumption of the second energy;
The second energy consumption calculating means determines which of the plurality of first segments belongs to the first energy consumption received as an input in comparison with the first threshold, The consumption amount of the second energy in the second period is calculated using the calculation formula set for the second segment corresponding to the determined first segment. 2. The heterogeneous energy consumption estimation apparatus according to 2. The consumption of the first energy of the first consumer selected from the aggregate and the identification information of the group to which the consumer belongs are accepted as input and set for each second segment. A second energy consumption calculation means for calculating the consumption amount of the second energy in the second period using the calculated equation;
The second energy consumption amount calculation means compares the first energy consumption amount received as an input with the first threshold value calculated for the group of the identification information received as an input. The second segment in the second period is determined using the calculation formula set for the second segment corresponding to the determined first segment. The energy consumption estimation apparatus according to claim 3, wherein the energy consumption is calculated. From the known consumption of the first energy in the first period for each consumer in a set of consumers consuming both different first energy and second energy, the same or different first energy consumption as the first period. A heterogeneous energy consumption estimation device that estimates the consumption of unknown second energy for two periods,
Storage means for storing one or more first threshold values that are boundary values for dividing the distribution range of the consumption amount of the first energy in the first period into a plurality of first segments;
The consumption of the first energy of the first period of one consumer selected from the aggregate is accepted as an input, and the consumption of the second energy of the second period using a predetermined calculation formula A second energy consumption calculating means for calculating
A pair of known first energy consumption of the first period and known second energy consumption of the second period for each consumer in the aggregate as individual sample values;
The first threshold corresponds to one or more second thresholds, each of which is a boundary value that divides the distribution range of the second energy consumption amount in the second period into a plurality of second segments, and the first segment The ratio of the number of sample values belonging to one corresponding second segment out of the total number of sample values belonging to one first segment with respect to all or part of a preselected portion A certain hit rate is set to be greater than or equal to a predetermined value,
The calculation formula is set for each second segment,
The second energy consumption calculating means determines which of the plurality of first segments belongs to the first energy consumption received as an input in comparison with the first threshold, Using the calculation formula set for the second segment corresponding to the determined first segment, the consumption amount of the second energy in the second period is calculated. Estimating device. From the known consumption of the first energy in the first period for each consumer in a set of consumers consuming both different first energy and second energy, the same or different first energy consumption as the first period. A heterogeneous energy consumption estimation device that estimates the consumption of unknown second energy for two periods,
The aggregate is divided into a plurality of groups based on one or more attributes of the consumer;
One or more first threshold values which are boundary values for dividing the distribution range of the consumption amount of the first energy in the first period into a plurality of first segments, and each of the consumers belongs to any of the plurality of groups Storage means for storing information that can be identified;
The consumption of the first energy of the first period selected from the aggregate and the identification information of the group to which the consumer belongs are accepted as input, and using a predetermined calculation formula, A second energy consumption calculating means for calculating a consumption amount of the second energy in the second period;
A pair of known first energy consumption of the first period and known second energy consumption of the second period for each consumer in the aggregate as individual sample values;
The first threshold corresponds to one or more second thresholds, each of which is a boundary value that divides the distribution range of the second energy consumption amount in the second period into a plurality of second segments, and the first segment The ratio of the number of sample values belonging to one corresponding second segment out of the total number of sample values belonging to one first segment with respect to all or part of a preselected portion It is set for each group so that a certain hit rate is above a predetermined value.
The calculation formula is set for each second segment,
The second energy consumption calculating means compares the first energy consumption received as an input with a plurality of the first threshold values set for the group of the identification information received as an input. The second segment in the second period is determined using the calculation formula set for the second segment corresponding to the determined first segment. A heterogeneous energy consumption estimation apparatus characterized by calculating energy consumption.   The first threshold value corresponds to one first segment in a range in which the hit ratio is equal to or greater than the predetermined value for each of the preselected portions of the first segment. The heterogeneous energy consumption estimation apparatus according to claim 6 or 7, wherein the number of hit points, which is the number of sample values belonging to both of two segments, is set to be maximum.   The calculation formula is a regression equation set for each second segment with the consumption amount of the second energy in the second period as an objective variable and the consumption amount of the first energy in the first period as an explanatory variable. The heterogeneous energy consumption estimation apparatus according to any one of claims 4 to 8, wherein   The multiple regression equation further comprising at least one of one or more attributes of the one consumer selected from the collection as an explanatory variable. Heterogeneous energy consumption estimation device.   The heterogeneous energy consumption estimation device according to claim 9 or 10, wherein the regression equation is a regression equation derived by regression analysis using the sample values belonging to the corresponding second segment.   The consumers are selected one by one from the aggregate, and the consumption of the first energy of the consumer for the first period is input to the second energy consumption calculation means, and the second energy consumption 12. The apparatus according to claim 4, further comprising a second energy consumption totalizing unit that totalizes the consumption amount of the second energy in the second period sequentially output from the amount calculating unit. Heterogeneous energy consumption estimation device. Known sample data in which an individual sample value is a pair of one first characteristic value and the other second characteristic value of the two events in a set of individuals exhibiting predetermined characteristics for two different events. A heterogeneous characteristic estimation device for estimating the unknown second characteristic value from the known first characteristic value based on:
Storage means for storing the sample data;
One or more first threshold values, which are boundary values for dividing the distribution range of the first characteristic value into a plurality of first segments, are boundary values for dividing the distribution range of the second characteristic value into a plurality of second segments. A first threshold value calculating means for calculating each of the second threshold values corresponding to one or more preset second threshold values,
The first threshold value calculation means corresponds to the first number corresponding to the whole number of the sample values belonging to one first segment for all of the first segment or a part selected in advance. The heterogeneous characteristic estimation apparatus according to claim 1, wherein the first threshold value is calculated so that a hit ratio that is a ratio of the number of sample values belonging to two segments is equal to or greater than a predetermined value.   The first threshold value calculation means corresponds to one first segment within a range in which the hit rate is equal to or greater than the predetermined value for each of the preselected portions of the first segment. 14. The heterogeneous characteristic estimation apparatus according to claim 13, wherein the first threshold value is calculated so that a target number that is the number of sample values belonging to both of the second segments is maximized. The aggregate is divided into a plurality of groups based on one or more attributes of the individual;
The storage means stores the sample data so that the individual sample values can be identified to which of the plurality of groups,
The second threshold value of 1 or more is set in common through a plurality of the groups,
The heterogeneous characteristic estimation apparatus according to claim 13 or 14, wherein the first threshold value calculating unit calculates the first threshold value corresponding to the second threshold value for each group.