EP2944891A1

EP2944891A1 - Room temperature estimating device, program

Info

Publication number: EP2944891A1
Application number: EP14738327.7A
Authority: EP
Inventors: Atsushi Mise; Naoki Muro; Keiichi Maruyama
Original assignee: Panasonic Intellectual Property Management Co Ltd
Current assignee: Panasonic Intellectual Property Management Co Ltd
Priority date: 2013-01-11
Filing date: 2014-01-09
Publication date: 2015-11-18
Anticipated expiration: 2034-01-09
Also published as: JP6160945B2; EP2944891A4; EP2944891B1; JP2014134360A; CN104919252A; CN104919252B; WO2014109290A1

Abstract

A room temperature estimating device (10) includes a storage (13), a prediction formula producer (15), a prediction change obtainer (16), and a room temperature estimator (17). The prediction formula producer (15) produces, based on pieces of room temperature data and pieces of outside air temperature data corresponding to a specified time in each of two or more days during a given extraction period, which are stored in the storage (13), a prediction formulae expressing a relation between the pieces of the room temperature data and the pieces of the outside air temperature data at the specified time. The room temperature estimator (17) determines, based on a temporal change in outside air temperature obtained by the prediction change obtainer (16), an outside air temperature at a target date and time corresponding to the specified time, and applies the determined outside air temperature to the prediction formula, and thereby to estimate a room temperature at the target date and time.

Description

TECHNICAL FIELD

The invention relates to a room temperature estimating device configured to estimate a room temperature at a target date and time, and a program that causes a computer to function as the room temperature estimating device.

BACKGROUND ART

There has been known a technique of adjusting a room temperature to a desired temperature at a scheduled time, based on information about temporal change in the room temperature and a predicted outside air temperature (see, e.g., JPH06-42765A, hereinafter referred to as "Document 1"). Also, regarding a temperature in a vehicle, there has also been known a technique of estimating a change in temperature of an in-vehicle space based on an estimated change in amount of solar radiation, a measured outside air temperature and a measured in-vehicle space temperature, and issuing a warning when the in-vehicle space temperature is predicted to reach a predetermined threshold (see, e.g., JP2005-343386A , hereinafter referred to as "Document 2").
Document 1 discloses a technique of measuring the room temperature as environmental information, and estimating the temporal change in the room temperature based on a history of the measured room temperatures. Document 1 also discloses a technique of determining an operation start time and a heating start time of a heater, based on the estimated temporal change in the room temperature and the outside air temperature, required amount of heat for room heating (hereafter, called "heating road energy"), and heating capability of the heater. Specifically, in Document 1, estimated values of the room temperature and the outside air temperature are determined, and the heating road energy for adjusting the room temperature to a desired temperature is calculated based on the estimated values.
In the configuration described in Document 1, the room temperature is estimated in order to calculate the heating road energy. However, in Document 1, the room temperature is estimated based on the history data of the room temperature. Document 1 does not disclose a technique for estimating the room temperature using another factor which the room temperature depends on.
Document 2 discloses a technique of measuring the room temperature and the outside air temperature as environmental information, and estimating a in-vehicle space temperature based on the measured in-vehicle space temperature and the measured outside air temperature along with the estimated amount of sun light.
The configuration described in Document 2 is directed to estimate the in-vehicle space temperature. It should be note that the in-vehicle space temperature follows the outside air temperature in short time when the outside air temperature changes. Therefore, it is relatively easy to estimate the in-vehicle space temperature based on the outside air temperature and the amount of sun light. On the other hand, the temperature in a building room depends on the thermal characteristics, such as heat insulating property, of the room, and does not immediately change when the outside air temperature changes. It is therefore difficult to estimate a room temperature in a building from the outside air temperature, based on the technique described in Document 2.
It has also been known a technique of estimating a room temperature in a building by use of a computer simulation, based on various factors such as the outside air temperature, the heat insulating property of the building, the sun light, the air ventilation, the rain fall, presence or absence of a person and the like. However, such kinds of computer simulation require a lot of information, and may further require dedicated measurement in order to obtain an accurate value. Therefore, this technique is not convenient for estimating the room temperature.

DISCLOSURE OF INVENTION

An object of the invention is to provide a room temperature estimating device for estimating a temperature of a room in a building based on measured environmental information, without complicated computer simulation, and to provide a program that causes a computer to function as the room temperature estimating device.
A room temperatures estimating device according to the invention includes: a room temperature obtainer configured to obtain room temperature data from a room temperature meter; an outside air temperature obtainer configured to obtain outside air temperature data from an outside air temperature meter; a storage configured to store the room temperature data obtained by the room temperature obtainer and the outside air temperature data obtained by the outside air temperature obtainer with each piece of the room temperature data and the outside air temperature data associated with a corresponding measured date and time; a prediction formula producer configured, based on pieces of the room temperature data and pieces of the outside air temperature data corresponding to a specified time in each of two or more days during a given extraction period, which are stored in the storage, to produce a prediction formulae expressing a relation between the pieces of the room temperature data and the pieces of the outside air temperature data at the specified time; a prediction change obtainer configured to obtain a predicted temporal change in outside air temperature; and a room temperature estimator configured, based on the temporal change in outside air temperature obtained by the prediction change obtainer, to apply an outside air temperature at a target time corresponding to the specified time to the prediction formula, and thereby to estimate a room temperature at the target time.
In the room temperature estimating device, preferably, the prediction formula producer is configured to produce, as the prediction formula, at least first and second prediction formulae respectively corresponding to at least first and second specified times. The first prediction formula is produced based on pieces of the room temperature data and pieces of the outside air temperature data corresponding to the first specified time in each of the two or more days during the given extraction period. The second prediction formula is produced based on pieces of the room temperature data and pieces of the outside air temperature data corresponding to the second specified time in each of the two or more days during the given extraction period. The room temperature estimator is configured to estimate a room temperature at a first target time corresponding to the first specified time by applying an outside air temperature at the first target time to the first prediction formula, and to estimate a room temperature at a second target time corresponding to the second specified time by applying an outside air temperature at the second target time to the second prediction formula.
In the room temperature estimating device, preferably, the prediction formula producer is configured to produce, as the prediction formula, a regression formula from pieces of the room temperature data and pieces of the outside air temperature data.
In the room temperature estimating device, preferably, the prediction formula producer is configured to produce the prediction formula by a simple linear regression analysis involving the outside air temperature data as an independent variable and the room temperature data as a dependent variable.
In the room temperature estimating device, preferably, the extraction period is included in any one of division periods which a period of one year is divided into based on climatic environment. The room temperature estimator is configured to apply, to prediction of a room temperature in a certain division period, a prediction formula produced based on pieces of the room temperature data and pieces of the outside air temperature data during an extraction period included in the certain division period.
Preferably, the room temperature estimating device further includes a correction information obtainer configured to obtain correction information corresponding to one state that has been selected from two or more states. The correction information is besides the outside air temperature and affects the room temperature. The prediction formula producer is configured to correct the prediction formula according to the state of the correction information obtained by the correction information obtainer, and thereby to produce a corrected prediction formula. The room temperature estimator is configured to estimate the room temperature based on the corrected prediction formula.
Preferably, the room temperature estimating device further includes an information outputter configured to transmit the room temperature estimated by the room temperature estimator to an informing device.
In the room temperature estimating device, preferably, the outside air temperature obtainer is configured to obtain outside air temperature data provided through a telecommunication network.
A program according to the invention is configured to cause a computer to function as the room temperature estimating device according to any of above described room temperature estimating devices.
With the configuration of the invention, it is possible to estimate a temperature of a room in a building based on easily measurable information, without complicated computer simulation.

BRIEF DESCRIPTION OF DRAWINGS

FIG. 1 is a block diagram illustrating Embodiment 1;
FIG. 2 is a graph for illustrating a principal of Embodiment 1;
FIG. 3 is a graph for illustrating the principal of Embodiment 1;
FIG. 4 is a block diagram illustrating Embodiment 2;
FIGS. 5A and 5B are graphs for illustrating a principle of Embodiment 2;
and
FIG. 6 is a block diagram illustrating Embodiment 3.

DESCRIPTION OF EMBODIMENTS

It will be explained a technique of estimating a temperature of a room in which air-conditioning is not performed, using an estimated temporal change in the outside air temperature. Under a condition where the air-conditioning is not performed, factors which the room temperature depends on include the outside air temperature, the heat insulating property of the room, the sun light (presence or absence of the sun light and the amount of sun light), the air ventilation (presence or absence of the air ventilation and the amount of air ventilation), the rain fall (whether it rains or not, and the amount of rainfall), the number of people in the room, and the like.
A heat insulating property of a room is unique characteristics for a residence, and it may be possible to roughly estimate the heat insulating property of the room based on the building materials of the residence, the construction method of the residence, and the like. However, it is not easy to quantitatively determine the heat insulating property of the room. Also, although it is possible to count the number of people in the room, it is difficult to theoretically determine the relation between the room temperature and the number of people in the room because the degree of influence on the increase in the room temperature differs between persons depending on their metabolic rate, amount of clothing, and the like. Also, it is possible to monitor the sun light, the air ventilation and the rain fall, but it is not easy to theoretically determine the influence thereof on the room temperature.
That is, it is possible to measure the factors which the room temperature depends on, but it is not easy to create a well-suited model that connects these factors with the room temperature. Therefore, it is not easy to obtain a room temperature from the measured values of these factors by a computer simulation. Also, a computer simulation which can estimate a room temperature with a practically needed accuracy requires so much information to be input and a correction processing. Therefore, estimating a room temperature for each room requires the professional's large work.
It is explained bellow a room temperature estimating device which can estimate a room temperature with a decent accuracy based on easily measurable information, without a complicated model-based computer simulation. In Embodiment 1, a technique of estimating a room temperature based only on an outside air temperature is explained. In Embodiment 2, a technique of estimating a room temperature based on an outside air temperature taking into account the heat insulating property of the room. In Embodiment 3, a technique of estimating a room temperature taking into account of influence of the sun light, the air ventilation, the rain fall and the number of people in the room.

(Embodiment 1)

A room temperature estimating device 10 of the present embodiment is configured to estimate a room temperature at a target date and time from an outside air temperature, based on a prediction formula that relates the outside air temperature and the room temperature. Therefore, the room temperature estimating device 10 includes a structure configured to produce the prediction formula and a structure configured to estimate a room temperature from an outside air temperature based on the prediction formula.
The room temperature estimating device 10 includes a device including a processor configured to execute programs to achieve below described functions, and a device for an interface, as main hardware components. The device including the processor may be a microcomputer with a built-in memory, a processor with an attached external memory, or the like. Also, a computer that executes programs for achieve the below described functions may function as the room temperature estimating device 10. Such kind of programs may be provided through a computer-readable storage medium, or be provided by communication via a telecommunication network.
First explained is the structure configured to produce the prediction formula in the room temperature estimating device 10. In order to produce the prediction formula, the room temperature and the outside air temperature are needed to be measured while being associated with respective dates and times. Therefore, as shown in FIG. 1, the room temperature estimating device 10 includes a room temperature obtainer 11 configured to obtain room temperature data (measured values) from a room temperature meter 21, and an outside air temperature obtainer 12 configured to obtain outside air temperature data (measured values) from an outside air temperature meter 22. The room temperature estimating device 10 further includes a storage 13, a clock 14, and a prediction formula producer 15. The storage 13 is configured to store the room temperature data (measured values) and the outside air temperature data (measured values) with each piece of the room temperature data and the outside air temperature data associated with a corresponding measured date and time. The clock 14 is configured to measure the current date and time. The prediction formula producer 15 is configured to produce two or more prediction formulae corresponding to respective two or more times of day.
The room temperature meter 21 is installed in a room of a building, and is configured to measure a temperature where the room temperature meter 21 is installed (i.e., measure a room temperature). The outside air temperature meter 22 is installed outside the building, and is configured to measure a temperature where the outside air temperature meter 22 is installed (i.e., measure an outside air temperature).
Each of the room temperature meter 21 and the outside air temperature meter 22 includes a temperature sensor configured to generate an analog output that reflects an ambient temperature, such as a thermistor, and a sensor amplifier configured to amplify the output of the temperature sensor. Each of the room temperature meter 21 and the outside air temperature meter 22 further includes a converter configured to convert the output of the sensor amplifier into digital data, and a communicator configured to transmit the digital data of the converter to the room temperature estimating device 10.

Each of the room temperature meter 21 and the outside air temperature meter 22 may not include the communicator or may not include the converter and the communicator. However, in view of transmitting measured values precisely to the room temperature estimating device 10, each of them desirably includes the converter and the communicator. In the case where the converter is not provided, the room temperature meter 21 and/or the outside air temperature meter 22 provide analog data to the room temperature estimating device 10.

Communication between the room temperature meter 21 or the outside air temperature meter 22 and the room temperature estimating device 10 is performed desirably through a wireless communication channel with a radio wave used as transmission medium, or through a wired communication channel. The room temperature meter 21 may share a casing with the room temperature estimating device 10. In the structure where the room temperature meter 21 shares the casing with the room temperature estimating device 10, the room temperature meter 21 is not necessarily to include the communicator.
The room temperature data (measured values) obtained by the room temperature obtainer 11 and the outside air temperature data (measured values) obtained by the outside air temperature obtainer 12 are stored in the storage 13 with each piece of the room temperature data and the outside air temperature data associated with a corresponding measured date and time. That is, the storage 13 is configured to store two kinds of sets of two information pieces, i.e., (room temperature, date and time) and (outside air temperature, date and time); or store sets of three information pieces, i.e., (room temperature, outside air temperature, date and time). The latter case is smaller in the data amount, and can save the capacity of the storage 13.
The date and time to be stored in the storage 13 is measured by the clock 14 provided in the room temperature estimating device 10. Dates and times at which the room temperature data and the outside air temperature data are to be obtained are preset in the room temperature obtainer 11 and the outside air temperature obtainer 12, respectively. The room temperature obtainer 11 and the outside air temperature obtainer 12 are configured to obtain room temperature data and outside air temperature data at each of the preset dates and times, respectively, based on the current date and time measured by the clock 14. In this configuration, it is desirable that the storage 13 is configured to store the sets of three information pieces, (room temperature, outside air temperature, date and time).
For example, each of the room temperature obtainer 11 and the outside air temperature obtainer 12 is configured to obtain data for each hour. For example, each of the room temperature obtainer 11 and the outside air temperature obtainer 12 is configured to obtain data at each hour. The room temperature obtainer 11 and the outside air temperature obtainer 12 is not necessarily to be configured to obtain data for each hour, but may obtain data for each 10 minutes, each 15 minutes, each 30 minutes, each two hours, or the like, one of which may be selected as needed. The shorter the time intervals are, the larger amount of information is obtained, which would increase the estimation accuracy of the prediction formula. However, this causes increase in the data amount to be stored in the storage 13. It is therefore preferable that the time intervals for obtaining data be set to a period around one hour, and be set in a range from a fraction of one hour to several hours. Each of the intervals for obtaining data by the room temperature obtainer 11 and the outside air temperature obtainer 12 is preferably set to a value obtained by dividing 24 hours by an integer.
Each of the room temperature meter 21 and the outside air temperature meter 22 may include a dedicated clock configured to measure the current date and time. In this configuration, the room temperature meter 21 and the outside air temperature meter 22 are configured to obtain the room temperature data and the outside air temperature data based on the dates and times measured by their own clocks, respectively, and transmit the obtained data to the room temperature estimating device 10. In other words, the room temperature meter 21 and the outside air temperature meter 22 are configured to transmit their respective room temperature data and outside air temperature data to the room temperature estimating device 10 with pieces of the room temperature data and the outside air temperature data associated with the corresponding date and time measured by their own clocks, respectively.
In this configuration, it is desirable that the storage 13 is configured to store the two kinds of sets of two information pieces, i.e., (room temperature, date and time) and (outside air temperature, date and time). Note that each of the room temperature meter 21 and the outside air temperature meter 22 is not limited to be configured to transmit the room temperature data or the outside air temperature data at that time the room temperature or the outside air temperature is measured, but may be configured to transmit a collection of data over a half day, or one day.
It is desirable that, in a case where there is a difference between a date and time at which a piece of the room temperature data is obtained and a date and time at which a piece of the outside air temperature data is obtained, if the difference is a half or less of an interval for obtaining pieces of data (e.g., one-tenth or less of an interval for obtaining pieces of data), these pieces of the respective data be regarded to be obtained at a same date and time and are associated with the same date and time.
Incidentally, in a case where a room temperature is kept out of influence of the sun light and fluctuation in outside air temperature is small, the incoming thermal energy to the room and the thermal energy released from the room would be balanced. Therefore, in this case, it is possible to formulate a hypothesis that outside air temperatures and room temperatures at the same time per day show a linear relation.
The inventors have measured room temperatures and outside air temperatures at two or more times per day during a relatively long period. Then, by analyzing graphically a relation between room temperatures and outside air temperatures for each of the two or more times, the inventors have found that the outside air temperatures and the room temperatures show a linear relation at a specified time, as shown in FIG. 2. That is, it has found that a room temperature at a specified time can be represented by a prediction formula of a linear function involving the outside air temperature as a variable, and that a room temperature can be estimated from an outside air temperature based on the prediction formula. Specifically, it has found that the outside air temperature and the room temperature measured at a first specified time in each of two or more days show a linear relation, and also the outside air temperature and the room temperature measured at a second specified time in each of the two or more days show a linear relation.
Therefore, in the room temperature estimating device 10 of the present embodiment, the prediction formula producer 15 is configured to produce the prediction formula based on the outside air temperature and the room temperature at a specified time in each of two or more days. The prediction formula producer 15 is configured to extract pieces of the room temperature data and pieces of the outside air temperature data corresponding to a specified time in each of two or more days during a given extraction period, which are stored in the storage 13, to produce a regression formula from the pieces of the room temperature data and the pieces of the outside air temperature data corresponding to the same time in each of two or more days, and to employ the regression formula as the prediction formula. Specifically, the room prediction formula producer 15 is configured, based on the pieces of the room temperature data and the pieces of the outside air temperature data corresponding to the specified time (the same time) in each of two or more days during the given extraction period, which are stored in the storage 13, to produce the prediction formulae expressing a relation between the pieces of the room temperature data and the pieces of the outside air temperature data at the specified time.
Since the prediction formula is expected to be written as a linear function of the outside air temperature, the pieces of the room temperature data and the pieces of the outside air temperature data for producing the prediction formula should include pieces of data for three or more days. Therefore, the extraction period should be three or more days, and be selected from a range of 15 to 90 days, for example. The lower limit of the range, 15 days, corresponds to one period of 24 seasons in the solar year (a half month), and the upper limit of the range, 90 days, corresponds to one season, i.e., the spring, the summer, the autumn, or the winter. The days of the period is an example, and may be 30 days (about one month), or may be one year if the outside air temperature is not so changed over year. The days for obtaining data for producing the prediction formula may include consecutive days, or may be discontinuous days. For example, the prediction formula may be produced based on the room temperature data and the outside air temperature data measured: every day; or every second day; or every week, over one or more years.
The prediction formula producer 15 is configured to write a prediction formula by a formula of "θ1(t)=α*θ2(t)+β", based on the finding that there is a linear relation between pieces of the room temperature data θ1(t) and pieces of the outside air temperature data θ2(t) corresponding to a target time "t" during the extraction period. For producing the prediction formula, a linear function is produced from the pieces of the room temperature data θ1(t) and the pieces of the outside air temperature data θ2(t), based on a known calculation method such as least-square method. That is, the prediction formula producer 15 is configured to produce a regression prediction formula from the pieces of the room temperature data θ1(t) and the pieces of the outside air temperature data θ2(t) corresponding to the target time (the specified time) during the extraction period.
The regression prediction formula involves the outside air temperature at a specified time as the explanatory variable and the room temperature at the specified time as the dependent variable. In other words, the prediction formula producer 15 is configured to produce the prediction formula by a simple linear regression analysis involving the outside air temperature data as the independent variable and the room temperature data as the dependent variable. The time of day (specified time) for producing the regression prediction formula is selected from a time period during which the room temperature is not affected by the sun light and relies only on the outside air temperature, but the variation of the outside air temperature is relatively gentle. The prediction formula producer 15 is configured to employ the produced regression prediction formula as the prediction formula for determining the room temperature from the outside air temperature.
The prediction formula producer 15 is configured to produce two or more regression prediction formulae that respectively correspond to two or more times of day. The prediction formula producer 15 is configured to produce the regression prediction formulae, as prediction formulae at the respective times of day. Specifically, the prediction formula producer 15 is configured to produce two or more prediction formulae that respectively correspond to two or more specified times of day. Each of the prediction formulae is produced based on pieces of the room temperature data and pieces of the outside air temperature data corresponding to a specified time. Specifically, the prediction formula producer 15 is configured to produce at least first and second prediction formulae respectively corresponding to at least first and second specified times. The first prediction formula is produced based on pieces of the room temperature data and pieces of the outside air temperature data corresponding to the first specified time in each of the two or more days during the extraction period. The second prediction formula is produced based on pieces of the room temperature data and pieces of the outside air temperature data corresponding to the second specified time in each of the two or more days during the extraction period.
The room temperature estimating device 10 produces the two or more prediction formulae respectively corresponding to the two or more times of day with the above described way, and then estimates a room temperature from an outside air temperature based on a prediction formula. Specifically, the room temperature estimating device 10 produces (at least) the first prediction formula corresponding to the first specified time and the second prediction formula corresponding to the second specified time. The room temperature estimating device 10 employs the first prediction formula for estimating a room temperature at a time corresponding to the first specified time, and employs the second prediction formula for estimating a room temperature at a time corresponding to the second specified time.
For example, the room temperature estimating device 10 produces the first prediction formula based on pieces of the room temperature data and pieces of the outside air temperature data obtained at 4 a.m. (first specified time) in each of two or more days, and produces the second prediction formula based on pieces of the room temperature data and pieces of the outside air temperature data obtained at 5 a.m. (second specified time) in each of the two or more days. The room temperature estimating device 10 employs the first prediction formula for estimating a room temperature at 4 a.m. (a time corresponding to the first specified time) of a certain day, and employs the second prediction formula for estimating a room temperature at 5 a.m. (a time corresponding to the second specified time) of a certain day.
The structure configured to estimate a room temperature from an outside air temperature in the room temperature estimating device 10 is described in detail below. The room temperature estimating device 10 includes a prediction change obtainer 16 and a room temperature estimator 17. The prediction change obtainer 16 is configured to obtain a predicted temporal change in outside air temperature, based on time series pieces of the outside air temperature data (measured values) obtained by the outside air temperature obtainer 12 from the outside air temperature meter 22. The room temperature estimator 17 is configured to estimate a room temperature using the temporal change in outside air temperature.
The prediction change obtainer 16 is configured to apply the time series pieces of the outside air temperature data to any of pre-registered two or more kinds of templates of temporal changes in outside air temperature, and to predict the temporal change in outside air temperature based on the applied template. The prediction change obtainer 16 is configured, when applying the time series pieces of the outside air temperature data to any of templates of temporal change in outside air temperature, to limit the templates to be applied, taking into account the weather and/or season of the day.
Regarding the predicted change in out side air temperature, instead of employing the outside air temperature data (measured values) obtained by the outside air temperature obtainer 12 from the outside air temperature meter 22, it is possible to employ a temporal change in outside air temperature obtained by the outside air temperature obtainer 12 via a telecommunication network such as internet. That is, the outside air temperature obtainer 12 may have a function configured to obtain outside air temperature data via a telecommunication network from a service provider that provides local weather information. In this configuration, the prediction change obtainer 16 employs the outside air temperature data obtained by the outside air temperature obtainer 12 from the service provider.
The outside air temperature data provided through the telecommunication network is the data related to a specific location in an area where a target room of which room temperature is to be estimated exists, and is not the outside air temperature corresponding to the target room. However, such the provided data can be expected to have linear relation with the outside air temperature of this room. Therefore, the room temperature estimator 17 is configured to correct, based on an actual measured value of the room temperature, a room temperature estimated using the provided outside air temperature. As a result, it is possible estimate a room temperature properly based on the outside air temperature data provided via the telecommunication network.
The room temperature estimator 17 determines, based on the predicted temporal change in outside air temperature obtained by the prediction change obtainer 16, an outside air temperature at a target date and time. Determined the outside air temperature, the room temperature estimator 17 estimates a room temperature by applying the determined outside air temperature to the prediction formula produced by the prediction formula producer 15. In short, the room temperature estimator 17 is configured to estimate a room temperature at a target date and time by: determining, based on the predicted temporal change in outside air temperature, an outside air temperature at the date and time of which room temperature is to be estimated; and applying the determined outside air temperature to the prediction formula.
The room temperature estimating device 10 desirably includes an information outputter 18 configured to transmit the room temperature estimated by the room temperature estimator 17 to an informing device 23. The informing device 23 may be a dedicated device including a display, or a device including a display and communication function, such as a smart phone, a tablet, and a personal computer. In a case of adopting such devices as the informing device 23, the information outputter 18 is configured to communicate with these devices. The informing device 23 may be provided integrally in a housing of the room temperature estimating device 10, as an informing device 23 in FIG. 1 shown by a broken line.
The room temperature estimated by the room temperature estimator 17 may, not only be notified of a user via the informing device 23, but be used for controlling a device that possibly affects the room temperature, such as a ventilating fan, an air conditioner, an electric shutter, an electric curtain and an electric window. In a case of controlling the heating and/or cooling operation of an air-conditioning apparatus(es) (air conditioner), by using a room temperature estimated based on the temporal change in outside air temperature, it is possible to determine a suitable timing at which the air-conditioning apparatus(es) should be turned off. As a result, it is possible to save energy consumed for air conditioning.
For example, in the summer, in a case where it is predicted that the room temperature can be kept in a comfortable level without operating a cooling apparatus due to decrease in the room temperature in the night and if it is determined a timing at which the cooling apparatus is to be turned off, it is possible to prevent the useless operation of the cooling apparatus to save energy. Similarly, in the winter, in a case where it is predicted that the room temperature can be kept in a comfortable level without operating a heating apparatus due to increase in the room temperature in the day time and if it is determined a timing at which the heating apparatus is to be turned off, it is possible to prevent the useless operation of the heating apparatus to save energy.
It would be easily supposed that a formula expressing the relation between pieces of outside air temperature data and pieces of the room temperature data varies according to the season. For example, in FIG. 2, data points in the left hand side show relations between the room temperature and the outside air temperature in the winter, and data points in the right hand side show relations between the room temperature and the outside air temperature in the summer. With a glance of the graph, it seems that data points in the left hand side group and data points in right hand side group can be represent by a single linear function. However, as shown in FIG. 3, by producing respective linear function from the points in the left hand side group (shown by squares in FIG. 3) and the points in the right hand side group (shown by triangles in FIG. 3), different prediction formulae (indicated by straight lines) are obtained between the groups.
It is therefore desirable that an extraction period, for measuring the room temperature and the outside air temperature used for producing the prediction formula, is determined for each season. In this configuration therefore, an extraction term is given for each division period which a period of one year is divided into. Desirably, a length of the division period corresponds to a period (a period derived from dividing a period of one year based on climatic environment) which is appropriately selected from a range of a quarter of one year to one-twenty fourth of one year (in the case of the "quarter of one year", the division periods reflect four seasons of spring, summer, autumn, and winter; and in the case of the "one-twenty fourth of one year", each division period corresponds to a half month). The length of a division period may be set to 15 to 90 days, and an extraction period for each division period may be three days or more. The division periods and their extraction periods are preliminarily stored in the storage 13, for example.
In an example, the prediction formula producer 15 is configured to produce a plurality of prediction formulae corresponding the number of division periods. Specifically, the prediction formula producer 15 is configured to produce two or more prediction formulae respectively corresponding to two or more times of day for each division period. The room temperature estimator 17 is configured, when predicting a room temperature at a certain time of a certain day, to select, from the two or more prediction formulae produced for each of the division periods, a prediction formula corresponding to a target time in a division period which a target date belongs to, and to estimate a room temperature from a temporal change in outside air temperature based on the selected prediction formula.
In another example, the prediction formula producer 15 is configured to newly produce two or more prediction formulae respectively corresponding to two or more times of day, at a transition from one division period to the next division period (for example, transition from a period of "the summer" to a period of "the autumn") determined based on the current date and time measured by the clock 14. After the prediction formulae are produced, the room temperature estimator 17 estimates a room temperature based on the newly produced prediction formulae.

(Embodiment 2)

In Embodiment 1, the prediction formula producer 15 produces prediction formulae each of which is based on the room temperature and the outside air temperature measured in a time period during which the room temperature is not affected by the sun light. Therefore, the time period which the prediction formulae are applicable to for estimating a room temperature from the outside air temperature is restricted. In other words, a room temperature in a time period during which the room temperature is affected by the sun light cannot be accurately estimated based on the prediction formulae produced according to Embodiment 1. The prediction formulae according to Embodiment 1 may be applicable only for a time period during which the change in outside air temperature is comparatively gentle, such as a period from the midnight to the early morning.
Explained in the present embodiment is a technique of determining a prediction formula applicable to a time period of day time during which the room temperature is affected by the sun light. Accordingly, the prediction formulae produced according to the technique of Embodiment 1 is employed for a period of the night time during which the influence of the sun light can be omitted, and the prediction formula explained below is employed for the day time during which the influence of the sun light should be considered. That is, in the present embodiment, different kinds of prediction formulae are employed for the time period during which the room temperature would not be affected by the sun light (i.e., a time period of no sun light) and for the time period during which the room temperature would be affected by the sun light. In the room temperature estimating device 10 of the present embodiment, a prediction formula producer 15 has a function configured to produce two kinds of prediction formulae.
As shown in FIG. 4, the room temperature estimating device 10 includes a first prediction formula producer 151 and a second prediction formula producer 152. The first prediction formula producer 151 is configured to produce prediction formulas (first kind of prediction formula) based on the same technique as that in Embodiment 1. The second prediction formula producer 152 is configured to produce a prediction formula (second kind of prediction formula) based on the technique described below.
The first prediction formula producer 151 is configured to produce the prediction formulae as is the prediction formula producer 15 in Embodiment 1. The first prediction formula producer 151 is configured to produce a regression prediction formula based on pieces of room temperature data and pieces of outside air temperature data corresponding to a specified time in each of two or more days, which are stored in the storage 13, and to employ the produced regression prediction formula as a prediction formula.
On the other hand, the second prediction formula producer 152 is configured to produce a prediction formula with the following scheme under an assumption that a relation between the room temperature and the outside air temperature depends on the thermal-characteristics (such as heat insulating property and heat storage performance) of the room. It will be assumed that the room temperature depends only on the outside air temperature, and it will be made a model in which the heat is transferred through partitions of the room such as walls, a ceiling and a floor, and the room temperature changes according to the change in the outside air temperature. In this model, the influence of the outside air temperature on the room temperature would vary according to the extent of the thermal conductivity of the partitions and the extent of the heat capacitance of the partitions. In the embodiment, the temperature of the air inside the room is regarded as the room temperature, and the radiant heat from the partitions is omitted.
According to the model described above, the room temperature would change later than the change in the outside air temperature. The inventors have reviewed results of experiments and found that there exist a particular relation between the change in the outside air temperature and the change in the room temperature, and that the room temperature changes later by a delay time than the change in the outside air temperature, and that the delay time depends on the thermal-characteristics (such as the heat insulating property and the heat storage performance) of the partitions. Furthermore, they have found, by determining the delay time, that a relation between a room temperature at a specific time and an outside air temperature at a time shifted by the delay time from the specific time can be expressed by a simple prediction formula, and that a room temperature at a desired dime can be estimated from an outside air temperature based on this prediction formula.
FIG. 5A is a graph showing points each of which is a relation between a room temperature and an outside air temperature measured at a same date and time. There seems no relationship between the room temperature and the outside air temperature with a glance of the graph. In contrast, as described above, the present embodiment is achieved based on the presumption that there is a correlation between the temporal change in room temperature and the temporal change in outside air temperature, if provided is a delay time which depends on the thermal-characteristics of the room.
Therefore, the room temperature estimating device 10 of the present embodiment includes an evaluator 19 configured to determine a time difference, based on pieces of room temperature data (measured values), pieces of outside air temperature data (measured values), and the associated dates and times, which are stored in the storage 13, so that a correlation coefficient between the room temperature and the outside air temperature becomes maximum. The evaluator 19 is configured to determine the time difference (hereinafter, also called "optimum time difference"), based on pieces of the room temperature data and pieces of the outside air temperature, during a certain focused term (hereinafter, called "extraction term"), by relatively shifting one of corresponding measured dates and times of the pieces of the room temperature data and corresponding measured dates and times of the pieces of the outside air temperature data so that a correlation coefficient between the pieces of the room temperature data and the pieces of the outside air temperature data becomes maximum. The extraction term is not limited to one day, but may be two or more days. In the example described below, the measured dates and times corresponding to the outside air temperature are shifted with respect to the measured dates and times corresponding to the room temperature. However, the contrary case is possible in which the measured dates and times corresponding to the room temperature are shifted with respect to the measured dates and times corresponding to the outside air temperature.
In this example, a piece of the room temperature data and a piece of the outside air temperature data corresponding to a certain date and time "t" are represented by "θ1(t)" and "θ2(t)", respectively. A data obtaining interval of pieces of room temperature data "θ1(t)", or pieces of outside air temperature data "θ2(t)", is represented by "p". A certain date and time "t" is represented by a formula "t=tθ+n*p", and a certain time difference "Δt" is represented by a formula "Δt=m*p where "t0" is a base point determined according to the "extraction term", and "m" and "n" are each a natural number.
According to the above notation, pieces of room temperature data are represented by "θ1(t0+p)", "θ1(t0+2p)", "θ1(t0+3p)", ..., and pieces of outside air temperature data are represented by "θ2(t0+p)", "θ2(t0+2p)", "θ2(t0+3p)", ... A piece of outside air temperature data corresponding to a date and time earlier by a time difference "Δt" than a date and time corresponding to a piece of room temperature data represented by "θ1(t0+n*p)" is represented by a formula "θ2(t0+n*p-Δt)= θ2(t0+(n-m)p)".
In the following, it is considered a period of a certain extraction term which is defined as "[t0+p, t0+q*p]". In this case, an average value "a(θ1)" of the pieces of the room temperature data θ1(t), during a period of the extraction term, is calculated as an average of pieces of the room temperature data, {θ1(t0+p), θ1(t0+2p), ..., θ1(t0+q*p)}. An average value "a(θ2)" of the pieces of the outside air temperature data θ2(t), during a period which is earlier by "the time difference Δt" than "the extraction term", is calculated as an average of pieces of the outside air temperature data, {θ2(t0+(1-m)p), θ2(t0+(2-m)p), ..., θ2(t0+(q-m)p)}. Where, [t0+p, t0+q*p] represents a certain closed interval, and includes discrete values of {t0+p, t0+2p, t0+3p, ..., t0+q*p}, the number of which is "q".
The evaluator 19 is configured, based on these values, to calculate a correlation coefficient between the pieces of the room temperature data θ1(t) corresponding to certain dates and times "t" and pieces of the outside air temperature data θ2(t-Δt) corresponding to dates and times "t-Δt", earlier by the time difference Δt(=m*p) than the certain dates and times. The correlation coefficient can be calculated by a known calculation method, and obtained by dividing a covariance of the pieces of data θ1(t) and the pieces of data θ2(t-Δt) by a product of a standard deviation of the pieces of the data θ1(t) and a standard deviation of the pieces of the data θ2(t-Δt). The abovementioned average values "a(θ1)" and "a(θ2)" are used for calculating the covariance and the standard deviations. The variable "t", representing the dates and times of the pieces of the room temperature data θ1(t) and the pieces of the outside air temperature data θ2(t-Δt), ranges within the period of the extraction term (i.e., the closed interval [t0+p, t0+q*p]).
The evaluator 19 is configured to calculate a correlation coefficient for each value of the number "m", while changing the value of the number "m" to change the time difference Δt. In the present embodiment, the maximum value of the number "m" is limited so that a product "m*p" of the number "m" and the time interval "p" does not exceed a length of one day. For example, in a case where the time interval "p" corresponds to one hour, the maximum value of the number "m" is limited so as not to exceed "24". The evaluator 19 is configured to determine a value "mm" of the number "m", corresponding to the maximum correlation coefficient. The evaluator 19 determines the optimum time difference "Δt_A" by a formula "Δt_A=mm*p".
FIG. 5B shows points each of which is a relation between a piece of the room temperature data θ1(t) and a piece of the outside air temperature data θ2(t-Δt_A) provided with the time difference (optimum time difference) Δt_A determined by the evaluator 19. In the illustrated example, it is possible to find that there is a linear relationship between the pieces of the room temperature data θ1(t) and the pieces of the outside air temperature data θ2(t-Δt_A), and that the relation between them can be expressed by a linear function.
The second prediction formula producer 152 is configured to produce, based on the relationship shown in FIG. 5B, a prediction formula for estimating a room temperature from an outside air temperature. The second prediction formula producer 152 is configured to extract, from the room temperature data and the outside air temperature data stored in the storage 13, pieces of the respective data during the extraction term, and to provide the time difference (the optimum time difference) Δt_A determined by the evaluator 19 to the dates and times corresponding to the extracted pieces of the outside air temperature data. Also, the second prediction formula producer 152 is configured, on the presumption that the pieces of the room temperature data θ1(t) and the pieces of the outside air temperature data θ2(t-Δt_A) have a linear relationship, to write the prediction formula by a formula of "θ1(t)=α*θ2(t-Δt_A)+β", and to determine the coefficients "α", "β" of the formula by a known calculation method such as least-square method. Specifically, the second prediction formula producer 152 produces the prediction formula by a simple linear regression analysis involving the outside air temperature data provided with the time difference Δt_A as the independent variable and the room temperature data as the dependent variable. In this way, the evaluator 19 determines the time difference Δt_A and the second prediction formula producer 152 determines the coefficients "α", "β", and as a result the prediction formula (second kind of prediction formula) can be produced. Note that the coefficients "α", "β" of the formula, in general, are different from the coefficients "α", "β" in the prediction formula produced by the first prediction formula producer 151.
That is, according to the second prediction formula producer 152, the evaluator 19 determines a time difference (optimum time difference), based on pieces of the room temperature data and pieces of the outside air temperature data, during a given extraction term, which are stored in the storage 13, and then the second prediction formula producer 152 produces a prediction formula, based on the pieces of the room temperature data and the pieces of the outside air temperature, the one of which the time difference is provided to.
The prediction formula produced by the second prediction formula producer 152 is applicable regardless of the influence of the sun light on the room temperature. Also, it is possible to estimate a room temperature by a single prediction formula regardless of the time of day. However, regarding a time period during which the room temperature is not affected by the sun light, it is possible to estimate a room temperature from an outside air temperature based on the prediction formula (first kind of prediction formula) produced by the first prediction formula producer 151, with a decent accuracy (possibly with an accuracy higher than that obtained by the prediction formula produced by the second prediction formula producer 152).
Therefore, it is desirable that the prediction formula produced by the first prediction formula producer 151 is employed for a time period of which room temperature can be estimated by the prediction formula produced by the first prediction formula producer 151, and that the prediction formula produced by the second prediction formula producer 152 is employed for the other time period so as to divide their responsibility. Specifically, the prediction formula (first kind of prediction formula) produced by the first prediction formula producer 151 is employed for the time period during which the room temperature would not be affected by the sun light (i.e. a time period of no sun light) and the prediction formula (second kind of prediction formula) produced by the second prediction formula producer 152 is employed for the time period during which the room temperature would be affected by the sun light.
In a case of estimating a room temperature from an outside air temperature based on the prediction formula produced by the second prediction formula producer 152, the room temperature estimating device 10 needs to obtain a piece of the outside air temperature data at a date and time which is earlier by the time difference (delay time) Δt_A determined by the evaluator 19 than a target date and time. Note that the "target date and time" is a date and time at which a room temperature is to be estimated.
Therefore, the room temperature estimator 17 determines, based on a predicted temporal change in outside air temperature obtained by a prediction change obtainer 16 and the time difference (optimum time difference) determined by the evaluator 19, an outside air temperature (a measured value or a prediction value) at a date and time earlier by the time difference than the target date and time. Determined the outside air temperature, the room temperature estimator 17 estimates a room temperature by applying the determined outside air temperature to the prediction formula produced by the prediction formula producer 15. In short, the room temperature estimator 17 is configured to determine, based on the predicted temporal change in outside air temperature, an outside air temperature at a time point earlier by the time difference determined by the evaluator 19 than a date and time of which room temperature is to be estimated; and to apply the determined outside air temperature to the prediction formula, and thereby to estimate a room temperature at a target date and time.
As described above, the prediction formula producer 15 of the present embodiment includes the first prediction formula producer 151 and the second prediction formula producer 152. The room temperature estimator 17 is configured to determine whether the current time is in the time period during which the room temperature is affected by the sun light or in the time period during which the room temperature is not affected by the sun light, based on the measured date and time of the clock 14. The prediction formula produced by the first prediction formula producer 151 is used for the time period during which the room temperature is not affected by the sun light, and the prediction formula produced by the second prediction formula producer 152 is used for the time period during which the room temperature is affected by the sun light.
As explained in Embodiment 1, the prediction formula produced by the first prediction formula producer 151 would vary according to the season. It is also easily supposed that the prediction formula produced by the second prediction formula producer 152 varies according to the season. It is therefore desirable that an extraction term, for measuring the room temperature and the outside air temperature used for producing the prediction formula, is determined for each season.
It is therefore defined division periods which a period of one year is divided into, and an extraction term is given for each division period. A length of the division period is appropriately selected from a range of a quarter of one year to one-twenty fourth of one year (in the case of the "quarter of one year", the division periods reflect four seasons of spring, summer, autumn, and winter; and in the case of the "one-twenty fourth of one year", each division period corresponds to a half month).
In this configuration, the second prediction formula producer 152 produces prediction formulae of which the number corresponds to the number of division periods. The room temperature estimator 17 selects, from the respective prediction formulae corresponding to the division periods, a prediction formula corresponding to a division period which a target date and time belongs to, and estimates a room temperature based on the selected prediction formula using a temporal change in outside air temperature.
It should be noted that the room-thermal-characteristics possibly varies across the ages. Therefore, desirably, the room temperature estimator 17 is configured, when estimating a room temperature, to employ a time difference which is determined for each division period. Specifically, it is desirable that the evaluator 19 newly determines a time difference (optimum time difference) Δt_A for each elapse of a division period, that the second prediction formula producer 152 newly produces a prediction formula (second kind of prediction formula) corresponding to the newly determined time difference Δt_A, and that the room temperature estimator 17 estimates a room temperature based on the newly produced prediction formula (second kind of prediction formula). However, the room temperature estimator 17 may be configured to estimate a room temperature based on a time difference determined regarding any of the division periods. It is also possible to estimate a room temperature based on an average of time differences determined regarding two or more division periods.
As described above, the room temperature estimator 17 in the present embodiment uses different kinds of prediction formulae according to whether the room temperature is affected by the sun light or not. Also, the outside air temperature to be applied to differs according to the kinds of the prediction formula. Accordingly, it is possible to increase the prediction accuracy of the room temperature. Other structures and operations are similar to those in Embodiment 1.

(Embodiment 3)

In Embodiment 1 and Embodiment 2, the room temperature estimating device 10 is configured to estimate a room temperature based only on an outside air temperature. However, as described above, when the air-conditioning is not performed, factors which the room temperature depends on include the sun light, the air ventilation, the rain fall, the number of people in the room. Note that, if the air conditioning is performed by an air-conditioning apparatus(es) that has a function of controlling the room temperature, the temperature in the room relies on the operation state of the air-conditioning apparatus(es), and accordingly it is not possible to predict the room temperature by a prediction formula. In the explanation below, therefore, it is assumed that the air-conditioning is not performed.
For taking into consideration of the information on the sun light, the air ventilation, the rain fall and the number of present people, besides the outside air temperature, it would be considered to create a model for relating the respective information with the room temperature, and to apply numerical values regarding the respective information to this model. However, because causal relationship between them is complicated, such a model requires a complicated computer simulation. As a result, such a model requires a lot of parameters to be input and causes severe processing load.
Therefore, in order to prevent the increase in the number of parameters and the processing load, the present embodiment treats each of the respective information as correction information, limits the number of possible states of each kind of correction information, and determines a prediction formula for each state of the correction information. In a case where there are two or more kinds of correction information, the prediction formula producer 15 divides, for each kind of correction information, the state of the correction information into two or more levels. Then the prediction formula producer 15 produces prediction formulae (corrected prediction formulae) each of which corresponds to a specific combination of levels of the two or more kinds of correction information. It will be explained an example in which the technical solution of the present embodiment is applied to the configuration of Embodiment 1 shown in FIG. 1, but it is possible to apply the technical solution of the present embodiment to the configuration of Embodiment 2.
In the present embodiment, two levels (present and absent) are defined for each of the sun light, the air ventilation, and the rain fall. Contrary, regarding the number of present people, it is assumed that the room temperature is increased by a predetermined temperature value (for example, 0.5°C) per one person. By simplifying the kind of correction information and by limiting the number of possible states for each kind of correction information, the number of combination of the respective correction information is finite and relatively small.
The prediction formula producer 15 is configured to determine a prediction formula according to a combination of respective states of two or more kinds of correction information. The number of people in the room is reflected only on the coefficient "β" in the prediction formula. Therefore, there is no need to produce different prediction formulae according to the number of people. Regarding the correction according to the number of people, the room temperature estimator 17 may be configured to add a product of the number of present people and the predetermined temperature value to a room temperature estimated by the prediction formula. In the above example, therefore, eight of prediction formulae are produced according to kinds of the correction information on the sun light, the air ventilation and the rain fall.
The prediction formula producer 15 is configured to correct the coefficients "α" and "β" of a prediction formula according to the states of the respective correction information, and thereby to produce a corrected prediction formula. For example, correction amounts of the coefficients "α" and "β" are associated with the states of the respective correction information and stored in the storage 13. In a case where one kind of correction information is in a certain state (for example, in a case where the air ventilation is present), the prediction formula producer 15 retrieves correction amounts of the coefficients "α" and "β" corresponding to this certain state from the storage 13, and applies the retrieved correction amounts to the coefficients "α" and "β" of the prediction formula and thereby to produce a corrected prediction formula.
As shown in FIG. 6, the room temperature estimating device 10 of the present embodiment includes a correction information obtainer 32 configured to obtain respective correction information from a sun light detector 33, an air ventilation detector 34, a rain fall detector 35 and a human counter 36.
The sun light detector 33 may include a photo detector such as a photo diode and a photo transistor, and a judging unit configured to compare an output of the photo detector with a threshold to determine the light amount. The influence of the sun light on the room depends on whether a curtain and/or a shutter is opened or closed. It is therefore desirable that the sun light detector 33 has a function configured to detect whether the curtain and/or the shutter is opened or closed.
The air ventilation detector 34 may be configured to detect whether a ventilation fan is operated or not, and/or to detect whether a window is opened or closed and/or to measure an airflow in the room. The rain fall detector 35 may be configured to collect rain water to measure the weight of the collected rain water per certain period, and/or to detect presence or absence of rain drops from an outside room image. The correction information on the rain fall may be obtained from information provided by a service provider through a telecommunication network such as internet. The human counter 36 may be configured to count the number of people in the room based on an in-room image.
The state of the correction information on the sun light, the air ventilation and the rain fall may be divided into three or more levels according to their degrees, instead of only two levels of "present" and "absent". The state of the sun light may be divided into four levels, e.g., "strong", "medium", "weak" and "very weak". Likewise, the state of air ventilation and/or the rain fall may be divided into three or more levels.
The room temperature estimator 17 corrects the prediction formula based on the correction information obtained by the correction information obtainer 32 to produce a corrected prediction formula, and estimates a room temperature, from an outside air temperature, based on the corrected prediction formula. Note that the correction amounts of the coefficients "α" and "β" according to each level of the sun light, the air ventilation, the rain fall and the number of people may be determined statistically based on actual measured values. Other structures and operations are similar to those in Embodiment 1 or Embodiment 2.

Claims

A room temperature estimating device comprising:
a room temperature obtainer configured to obtain room temperature data;

an outside air temperature obtainer configured to obtain outside air temperature data;

a storage configured to store the room temperature data obtained by the room temperature obtainer and the outside air temperature data obtained by the outside air temperature obtainer with each piece of the room temperature data and the outside air temperature data associated with a corresponding measured date and time;

a prediction formula producer configured, based on pieces of the room temperature data and pieces of the outside air temperature data corresponding to a specified time in each of two or more days during a given extraction period, which are stored in the storage, to produce a prediction formulae expressing a relation between the pieces of the room temperature data and the pieces of the outside air temperature data at the specified time;

a prediction change obtainer configured to obtain a predicted temporal change in outside air temperature; and

a room temperature estimator configured, based on the temporal change in outside air temperature obtained by the prediction change obtainer, to apply an outside air temperature at a target time corresponding to the specified time to the prediction formula, and thereby to estimate a room temperature at the target time.
The room temperature estimating device according to claim 1,
wherein
the prediction formula producer is configured to produce, as the prediction formula, at least first and second prediction formulae respectively corresponding to at least first and second specified times, the first prediction formula being produced based on pieces of the room temperature data and pieces of the outside air temperature data corresponding to the first specified time in each of the two or more days during the given extraction period, the second prediction formula being produced based on pieces of the room temperature data and pieces of the outside air temperature data corresponding to the second specified time in each of the two or more days during the given extraction period, and
the room temperature estimator is configured
to apply an outside air temperature at a first target time corresponding to the first specified time to the first prediction formula, and thereby to estimate a room temperature at the first target time, and
to apply an outside air temperature at a second target time corresponding to the second specified time to the second prediction formula, and thereby to estimate a room temperature at the second target time.
The room temperature estimating device according to claim 1 or 2, wherein the prediction formula producer is configured to produce, as the prediction formula, a regression formula from pieces of the room temperature data and pieces of the outside air temperature data.
The room temperature estimating device according to any one of claims 1 to 3, wherein
the extraction period is included in any one of division periods which a period of one year is divided into based on climatic environment, and
the room temperature estimator is configured to apply, to prediction of a room temperature in a certain division period, a prediction formula produced based on pieces of the room temperature data and pieces of the outside air temperature data during an extraction period included in the certain division period.
The room temperature estimating device according to any one of claims 1 to 4, further comprising a correction information obtainer configured to obtain correction information corresponding to one state that has been selected from two or more states, the correction information is besides the outside air temperature and affects the room temperature, wherein
the prediction formula producer is configured to correct the prediction formula according to the state of the correction information obtained by the correction information obtainer, and thereby to produce a corrected prediction formula, and
the room temperature estimator is configured to estimate the room temperature based on the corrected prediction formula.
The room temperature estimating device according to any one of claims 1 to 5, further comprising an information outputter configured to transmit the room temperature estimated by the room temperature estimator to an informing device.
The room temperature estimating device according to any one of claims 1 to 6, wherein the outside air temperature obtainer is configured to obtain outside air temperature data provided through a telecommunication network.
The room temperature estimating device according to any one of claims 1 to 7, wherein the prediction formula producer is configured to produce the prediction formula by a simple linear regression analysis involving the outside air temperature data as an independent variable and the room temperature data as a dependent variable.
A program configured to cause a computer to function as the room temperature estimating device according to any one of claims 1 to 8.