JP4066370B2 - Method and apparatus for estimating generation of large amount of heat - Google Patents

Description

  The present invention relates to a method and apparatus for estimating the generation of a large amount of heat. More specifically, the present invention relates to a method and apparatus for estimating the occurrence time of large heat generation such as cooking.

  There is a conventional technique in which a temperature sensor and an illuminance sensor are installed in a room, and a occupant's occupancy status is estimated based on temporal changes in measured values from these sensors (Non-Patent Document 1).

Tomoko Wakasugi, Hirofumi Honma, Michiko Miwa, Osamu Yamamoto, Naoko Yoshimura, Hiromi Mohri: Research on type of living using environmental sensors, Abstracts of Annual Conference of Architectural Institute of Japan E2, pp 305-308 (5620), 1995 August

  In recent years, the number of highly airtight and highly insulated houses is increasing, and some houses adopt a duct-type whole building air conditioning system that can achieve both planned ventilation and indoor air conditioning. In order to realize energy saving operation of the entire building air conditioning system, it is desirable to perform air conditioning that predicts resident behavior, such as starting pre-cooling operation before cooking before the heat load increases rapidly.

  However, the technology of Non-Patent Document 1 cannot estimate the occurrence time of a large amount of heat generation such as cooking only by estimating the occupant's occupancy status. Further, in the technique of Non-Patent Document 1, since it is necessary to install an illuminance sensor in addition to the temperature sensor, there is a problem that an additional cost of the illuminance sensor is required. Technology that determines the occupant's occupancy requires a temperature sensor to detect subtle changes in temperature, so it is difficult to make a reliable determination only with the temperature sensor, and as a result, information other than temperature is obtained. Need a sensor for.

  Therefore, an object of the present invention is to provide a method and apparatus that can estimate the occurrence time of a large amount of heat generation factors such as cooking without using a sensor other than a temperature sensor.

In order to achieve such an object, the method for estimating generation of a large amount of heat according to claims 1 and 3 records room temperature information from a temperature sensor installed in a target room and time information corresponding to the room temperature information, and records the recorded information. Based on the obtained information, a temperature difference between the highest room temperature and the lowest room temperature in a predetermined time interval and a regression coefficient derived from a time change of the room temperature in the time interval are obtained, and one of the temperature difference and the regression coefficient or On condition that both are equal to or greater than a predetermined reference value for each, in a time zone corresponding to the time interval in which this condition is satisfied, in the target room than in the target room or a room other than the target room. It is estimated that a large fever has occurred.

  Therefore, by using a temperature difference between the highest room temperature and the lowest room temperature in a predetermined time interval range and a regression coefficient derived from the time change of the room temperature in the time interval, a rapid increase in room temperature can be captured. Thereby, it is possible to estimate the time zone in which a large amount of heat generation factors such as cooking occur, without using a sensor other than the temperature sensor.

  According to the first aspect of the present invention, a cumulative frequency line of the temperature difference is obtained, and a value corresponding to a point at which the slope of the cumulative power line is equal to or less than a predetermined value or a change amount in which the slope exceeds a predetermined value. A value corresponding to the smaller point is determined as a reference value for the temperature difference.

  The rapid rise in room temperature during cooking is very different from room temperature fluctuations in other time zones, and the occurrence frequency is low. For this reason, the cumulative frequency distribution draws a gentle gentle slope part after drawing a steep slope part to the right. A sudden rise in room temperature, such as during cooking, corresponds to a gentle slope. Therefore, an appropriate reference value can be set by setting a value corresponding to a point at which the steeply inclined portion transitions to the gently inclined portion as the reference value.

  The invention according to claims 2 and 3 obtains a cumulative frequency line of the regression coefficient, and a value corresponding to a point at which the slope of the cumulative frequency line is not more than a predetermined value or a change in which the slope exceeds a predetermined value. A value corresponding to a point that becomes smaller in quantity is determined as a reference value for the regression coefficient.

  The rapid rise in room temperature during cooking is very different from room temperature fluctuations in other time zones, and the occurrence frequency is low. For this reason, the cumulative frequency distribution draws a gentle gentle slope part after drawing a steep slope part to the right. A sudden rise in room temperature, such as during cooking, corresponds to a gentle slope. Therefore, an appropriate reference value can be set by setting a value corresponding to a point at which the steeply inclined portion transitions to the gently inclined portion as the reference value.

According to a fourth aspect of the present invention, in the method for estimating the generation of a large amount of heat according to any one of the first to third aspects, the target room is a kitchen, and cooking is started at a time when the large heat generation is estimated to occur. The time is estimated.
The calorific value generated in the kitchen during cooking is, for example, equal to or more than the calorific value required for heating the entire house, in other words, several times to 10 times the calorific value due to heating or the like normally generated in a room other than the kitchen, Very big. Therefore, it can be estimated that the time when the large heat generation that protrudes is the cooking start time. And during cooking, since a large amount of heat load is generated in a short time, the kitchen room temperature rises rapidly. By using a temperature difference between the maximum room temperature and the minimum room temperature in the range of a predetermined time interval and a regression coefficient derived from a time change of the kitchen room temperature in the time interval, it is possible to capture the rapid increase in the room temperature.

The invention according to claim 5 is the method for estimating generation of a large amount of heat according to any one of claims 1 to 4, wherein the target room is a room of a building including a whole building air conditioning system, and the time interval In addition to the above condition, the correlation coefficient between the room temperature of the target room and the room temperature of the other room is equal to or greater than a predetermined reference value.

  When the room temperature rise of the target room is large, the air warmed in the target room returns to the air conditioner and is sent to the other room through the duct, so that the room temperature of the room other than the kitchen rises in the same manner. Therefore, more accurate estimation is possible by using the correlation coefficient between the room temperature of the target room and the room temperature of the other room.

The apparatus for estimating generation of a large amount of heat according to claim 6 is an information recording means for recording room temperature information from a temperature sensor installed in each room of a building equipped with a whole building air conditioning system and time information corresponding to the room temperature information. And a regression derived from a temperature difference between the highest room temperature and the lowest room temperature of the target room at a predetermined time interval and a time change of the room temperature of the target room at the time interval based on the information recorded in the information recording means. A calculating means for obtaining a part or all of a coefficient and a correlation coefficient between the room temperature of the target room and the room temperature of the other room in the time interval, the temperature difference, the regression coefficient, and the correlation coefficient Comparing means for determining whether or not this condition is satisfied on the condition that any or all of these are equal to or greater than a predetermined reference value for each, the reference value of the temperature difference is the cumulative of the temperature difference Frequency line A value corresponding to a point where the slope of the cumulative frequency line is equal to or less than a predetermined value, or a value corresponding to a point where the slope decreases with a change amount exceeding a predetermined value, and the reference value of the regression coefficient is the regression A cumulative frequency line of coefficients is obtained, and is a value corresponding to a point at which the slope of the cumulative frequency line is equal to or less than a predetermined value or a value corresponding to a point at which the slope decreases with a change amount exceeding a predetermined value, and the condition is satisfied It is estimated that cooking is performed in the target room or large heat generation similar to cooking occurs in the time zone corresponding to the time interval.

According to the method for estimating the generation of a large amount of heat and the apparatus for estimating the generation of a large amount of heat according to the present invention , it is possible to estimate a time zone in which a large amount of heat generation factors such as cooking are generated with a simple configuration that does not use sensors other than temperature sensors. By estimating the cooking start time, the occupant's cooking start pattern can be reflected in the air conditioning. For example, it is possible to perform air-conditioning that predicts resident behavior, such as starting pre-cooling operation before cooking where the heat load increases rapidly, and energy-saving operation such as the entire building air-conditioning system can be realized. In addition, the estimation result of the cooking start time, in other words, information on whether cooking was performed in a relatively regular time zone is effective as one information for determining whether the resident is living normally. It can be used as a safety confirmation system. Furthermore, in addition to cooking, a large amount of heat generation, such as heat generation due to a fire, can be detected, and the time of occurrence can also be estimated.

Furthermore, according to the method for estimating the occurrence of a large amount of heat according to claim 5 , it is possible to estimate with higher accuracy by using the correlation coefficient between the room temperature of the target room and the room temperature of the other room. Moreover, it is not necessary to install a temperature sensor anew, and the room temperature detection function of each room originally provided in the entire building air conditioning system can be used as it is.

Furthermore, according to the method for estimating the generation of a large amount of heat according to claims 1 to 3 and the apparatus for estimating the generation of a large amount of heat according to claim 6 , it is possible to set a reference value for an appropriate temperature difference / regression coefficient, and to estimate with higher accuracy. Is possible.

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

  1 to 8 show an embodiment of the method and apparatus for estimating the generation of a large amount of heat according to the present invention. This method of estimating the generation of a large amount of heat records room temperature information from the temperature sensor 3 installed in the target room 1a and time information corresponding to the room temperature information, and sets a predetermined time based on the recorded information. A temperature difference ΔT between the maximum room temperature and the minimum room temperature in the interval Δt and a regression coefficient α derived from a time change of the room temperature in the time interval Δt are obtained, and both the temperature difference ΔT and the regression coefficient α are predetermined reference values. On the condition that it becomes Tx, αx or more, in the time zone corresponding to the time interval Δt in which this condition is satisfied, the heat generation is larger in the target room 1a than in the target room 1a or outside the target room 1a. Is estimated to have occurred.

  In particular, in the present embodiment, the target room 1a is a kitchen, and the time when the large heat generation is estimated is estimated as the cooking start time. The amount of heat generated in the kitchen 1a during cooking is, for example, equal to or more than the amount of heat required for heating the entire house, in other words, several times to 10 times the amount of heat generated by heating or the like that can normally occur in the room 1b other than the kitchen 1a. That's a big deal. Therefore, it can be estimated that the time at which the large heat generation is generated is the cooking start time.

  During cooking, a large amount of heat load is generated in a short time, so the kitchen room temperature rises rapidly. In order to catch this rapid rise in room temperature, a temperature difference ΔT between the highest room temperature and the lowest room temperature in a predetermined time interval Δt is used. Here, when the minimum room temperature is simply subtracted from the maximum room temperature, it becomes difficult to distinguish whether the temperature is rising or falling. Therefore, the regression coefficient α of the kitchen room temperature within the time interval Δt is also used. The regression coefficient α is obtained, for example, by linearly approximating the kitchen room temperature distribution within the time interval Δt by the least square method.

  Further, in the present embodiment, the target room, that is, the kitchen 1a is assumed to be a room of a building equipped with the entire building air conditioning system. For example, the building in this embodiment is a detached house having a total of five rooms 1 including a kitchen, a living room, a couple's bedroom, a children's room, and a Japanese-style room. The air conditioner 4 performs air conditioning continuously for 24 hours. Here, when the room temperature rise of the kitchen 1a is large, the room temperature of the room 1b other than the kitchen 1a rises similarly. This is because the air heated in the kitchen 1a returns to the air conditioner 4 and is sent to the other chamber 1b through the duct 6. Therefore, in this embodiment, the correlation coefficient r of the room temperature between the kitchen 1a and the other room 1b is also used for estimating the cooking start time. In this embodiment in which there are four rooms 1b other than the kitchen 1a and four correlation coefficients r are obtained from one time interval Δt, for example, an average value r_ave of these correlation coefficients r is used. .

  Each room 1 of the building provided with the entire building air conditioning system is provided with a room controller 2, and each of these room controllers 2 is provided with a temperature sensor 3. Each room controller 2 measures the room temperature of each room 1 at intervals of 1 minute and 1/3 ° C., for example, and transfers the measured room temperature information to the control unit 5 of the air conditioner 4. For example, in this embodiment, the external output function of the control unit 5 of the air conditioner 4 is used to record the room temperature information sent from each room controller 2 and the measurement time of the room temperature in association with each other. Thereby, it is not necessary to newly install the temperature sensor 3 in each room 1, and the room temperature detection function of each room 1 originally provided in the entire building air conditioning system can be used as it is.

  FIG. 3 shows an example of a relationship between room temperature information and time information sent from each room controller 2, that is, an example of a time change of each room temperature. The graph shown by the symbol A in FIG. 3 shows the time change of the room temperature in the living room, the graph shown by the symbol B shows the time change of the room temperature of the kitchen, the graph shown by the symbol C shows the time change of the room temperature of the couple bedroom, A graph indicated by a symbol D indicates a time change of the room temperature of the child room, and a graph indicated by a symbol E indicates a time change of the room temperature of the Japanese room. Here, in the present embodiment, the time interval Δt for obtaining the temperature difference ΔT, the regression coefficient α, and the correlation coefficient r described above is 10 before and after the time t (hereinafter also referred to as the calculation reference time t) serving as a calculation reference. Minutes, that is, 20 minutes are set. The calculation reference time t is advanced every minute. Therefore, for example, when the calculation reference time t is 7: 3, the temperature difference ΔT, the regression coefficient α, and the correlation coefficient r are obtained based on the room temperature information measured from 6:53 to 7:13. . For example, FIG. 4 shows the kitchen room temperature measured over 20 minutes from 6:53 to 7:13. The straight line 21 in the figure is a linear approximation of the kitchen room temperature distribution from 6:53 to 7:13 by the least square method, and the slope of this straight line 21 indicates that the calculation reference time t is 7: 3. Is the regression coefficient α. FIG. 5 shows the correlation between the room temperature of the kitchen from 6:53 to 7:13, and the room temperature of the other room 1b, (A) shows the correlation with the room temperature of the living room, and (B) shows the room of the couple's bedroom. The correlation with room temperature is shown, (C) shows the correlation with the room temperature of the child room, and (D) shows the correlation with the room temperature of the Japanese room. The time increment for advancing the calculation reference time t is called a calculation time step Δt ′, and the time step for measuring each room temperature is called a measurement time step Δt ″. In this embodiment, the measurement time step Δt ″ and the calculation time step Although Δt ′ is 1 minute and the time interval Δt is 20 minutes, these values are not necessarily limited.

  For example, in this embodiment, cooking is started at a calculation reference time t at which the temperature difference ΔT is equal to or greater than the reference value Tx, the regression coefficient α is equal to or greater than the reference value αx, and the average value r_ave of the correlation coefficient r is equal to or greater than the reference value Rx. The time is estimated. Thus, the misjudgment of cooking start time can be prevented by taking the logical product of three conditions.

  Here, in the present embodiment, the reference value Tx of the temperature difference ΔT and the reference value αx of the regression coefficient α are determined based on the cumulative frequency distribution. Specifically, the kitchen room temperature is recorded for a certain period, the temperature difference ΔT and the regression coefficient α are obtained based on the record, and the cumulative frequency distributions of the temperature difference ΔT and the regression coefficient α are obtained. Then, the temperature difference ΔT and the regression coefficient α corresponding to the transition point from the steeply inclined portion to the gently inclined portion in each cumulative power line are determined as reference values Tx and αx, respectively. Here, the cumulative frequency distribution may be a cumulative relative frequency distribution. Further, the cumulative frequency line may be a cumulative frequency line broken line, or may be an approximated cumulative frequency line broken line or a cumulative frequency distribution curve.

  The rise in room temperature during cooking is significantly different from room temperature fluctuations in other time zones, and the frequency of occurrence is low, about several times a day. For this reason, in the cumulative relative frequency distribution of the temperature difference ΔT and the regression coefficient α, a steeply sloped part is drawn upward and then a gentle slope part is drawn near 100%. It is considered that the value of the temperature difference ΔT and the regression coefficient α during cooking corresponds to the gentle slope portion. Therefore, it is possible to determine whether or not the temperature difference ΔT and the regression coefficient α are during cooking by determining the values that serve as the boundary between the steeply inclined portion and the gently inclined portion as the reference values Tx and αx. As a method for finding the boundary point between the steeply inclined portion and the gently inclined portion, for example, the slope of the tangent line is obtained from the left side to the right side of the cumulative frequency line, and the point where the slope is below a certain value is determined. For example, a method may be used in which a boundary point is used, or a point that has become smaller by an amount of change exceeding a certain value is used as the boundary point.

  FIG. 6 shows an example of the cumulative relative frequency distribution of the room temperature regression coefficient α of the kitchen 1a, and FIG. 7 shows an example of the cumulative relative frequency distribution of the temperature difference ΔT of the kitchen 1a. In the cumulative relative frequency distribution of the regression coefficient α in FIG. 6, the slope is gentle in the vicinity of “0.4 ° C./10 minutes”. Therefore, in this example, it is preferable to set the reference value αx of the regression coefficient α to “0.4 ° C./10 minutes”. Further, in the cumulative relative frequency distribution of the temperature difference ΔT in FIG. 7, for example, a part is stepped like a section of 0.8 ° C. to 1 ° C. difference. This is because the room temperature is recorded in 1/3 ° C. increments in this embodiment. Therefore, if the temperature increment becomes fine, the cumulative relative frequency line of the temperature difference ΔT becomes smooth. Assuming that the cumulative relative frequency distribution of the temperature difference ΔT in FIG. 7 is smoothly distributed, the slope becomes gentle in the vicinity of the “1.3 ° C. difference”. Therefore, in this example, it is preferable to set the reference value Tx of the temperature difference ΔT to “1.3 ° C. difference”.

  On the other hand, FIG. 8 shows an example of the cumulative relative frequency distribution of the average value r_ave of the room temperature correlation coefficient r between the kitchen 1a and the other room 1b. In the figure, there is almost no portion where the slope becomes gentle, and conversely, the slope is steep in the vicinity where the average value r_ave is 0.95. This is considered to be due to the fact that since the upper limit of the correlation coefficient r is 1, the value near 1 is very difficult to appear. It is difficult to determine the reference value Rx of the correlation coefficient r based only on this cumulative frequency distribution. Thus, for example, in the present embodiment, in the cumulative relative frequency distribution of the average value r_ave of the room temperature correlation coefficient r, the same degree as the cumulative relative frequency of the reference value Tx of the temperature difference ΔT or the cumulative relative frequency of the reference value αx of the regression coefficient α. Is set to the reference value Rx.

  However, the method of setting the reference values Tx, αx, Rx is not necessarily limited to the above example. For example, the temperature difference between the kitchen 1a before cooking and after cooking, the regression coefficient of the time change of the kitchen room temperature from the start of cooking to the end of cooking, and the correlation coefficient between the room temperature of the kitchen and the other room 1b during cooking And the reference values Tx, αx, Rx may be set based on these actually measured values.

  The method for estimating the generation of a large amount of heat described above can be implemented as an apparatus, for example, as shown in FIG. This apparatus for estimating the generation of a large amount of heat includes information recording means 10 for recording room temperature information from temperature sensors 3 installed in each room 1 of a building equipped with a whole building air conditioning system and time information corresponding to the room temperature information, Based on the information recorded in the recording means 10, a temperature difference calculation unit 11a that obtains a temperature difference ΔT between the maximum room temperature and the minimum room temperature of the kitchen 1a in the time interval Δt, and derived from the time change of the room temperature of the kitchen 1a in the time interval Δt. A calculation means 11 having a regression coefficient calculation unit 11b for calculating the regression coefficient α, and a correlation coefficient calculation unit 11c for calculating an average value r_ave of the correlation coefficient r between the room temperature of the kitchen and the room temperature of the other room 1b in the time interval Δt; Comparing means 12 for determining whether or not the condition of temperature difference ΔT ≧ reference value Tx, regression coefficient α ≧ reference value αx, and average value r_ave of correlation coefficient r ≧ reference value Rx is satisfied, and the condition And an output unit 13 for outputting the calculated reference time t as the estimated cooking time starts. For example, a known device such as a hard disk may be used as the information recording unit 10, and a known device such as a CPU may be used as the computing unit 11 and the comparison unit 12.

  FIG. 2 shows an example of processing executed by the estimation apparatus. In this process, first, the external output function of the control unit 5 of the air conditioner 4 is used, and the room temperature information of each room sent from each room controller 2 and the measurement time of the room temperature are recorded for a certain period of time. 10 (S1). The start time of the certain period is set as the measurement start time, and the end time is set as the measurement end time. The measurement start time and the measurement end time include date information. The reference values Tx, αx, Rx may be set using the information obtained in S1. Next, the initial time is set as the calculation reference time t (S2). For example, a time advanced by a half of the time interval Δt from the measurement start time is set as the initial time. Then, the following processing is performed until the calculation reference time t exceeds the end time (S3; Yes). Based on room temperature information from time interval Δt, that is, from time t−Δt / 2 to time t + Δt / 2, an average value r_ave of temperature difference ΔT, regression coefficient α, and correlation coefficient r is obtained (S4, S5, S6). Then, it is determined whether or not the condition of temperature difference ΔT ≧ reference value Tx and regression coefficient α ≧ reference value αx and average value r_ave of correlation coefficient r ≧ reference value Rx is satisfied (S7). When satisfying (S7; Yes), the calculation reference time t is recorded as the estimated cooking start time (S8). Then, the calculation reference time t is advanced by the calculation time increment Δt ′ (S9). Until the calculation reference time t exceeds the end time, the processing from S3 is repeated. For example, a time that is one-half of the time interval Δt from the measurement end time is set as the end time.

  After the above process is completed, the output unit 13 outputs the recorded estimated cooking start time. At this time, the estimated cooking start time within a certain time from the previous estimated cooking start time, for example, within one hour, is considered to belong to a series of cooking actions, and may be excluded. Here, for example, a display device such as a display may be used as the output unit 13 to display the estimation result to an operator, an analyst, or the like. The output result from the output unit 13 will be described below. It is also possible to use it.

  For example, the output unit 13 outputs the estimated cooking start time to the control unit 5 of the air conditioner 4. In the air conditioner 4, for example, based on the notified estimated cooking start time, it is possible to perform an energy saving operation that predicts resident behavior, such as starting a precooling operation before the predicted cooking start time. Further, in this case, the output unit 13 may output the day and month information of the estimated cooking start time to the control unit 5 of the air conditioner 4 together. In this case, the controller 5 of the air conditioner 4 can predict the cooking start time according to the day of the week or the month by learning the pattern of the cooking start time according to the day of the week or the month.

  Further, for example, the output unit 13 may output the estimated cooking start time to another system, for example, a safety confirmation system that remotely confirms whether the resident is living normally. Whether cooking is performed regularly is considered to be effective as information for determining whether, for example, an elderly person living alone lives normally.

  The room, room information and time information of each room 1 were recorded over a period of 9 days for a detached house with a total of 5 rooms: a kitchen, a living room, a couple's bedroom, a children's room, and a Japanese-style room. In addition, in order to confirm the operating condition of the air conditioner 4, the air volume of the fan of the air conditioner 4 was also recorded. The room temperature of the kitchen and the air volume of the fan are shown in FIGS. The code | symbol F in FIGS. 10-18 shows the time change of the kitchen temperature, and the code | symbol G shows the time change of the air volume of a fan. During the above period, there was a time zone in which the air conditioner 4 was temporarily stopped due to a trouble or the like (indicated by the symbol H in FIGS. 13 and 17), but the entire building air conditioning system was in a cooling operation almost continuously. I was doing.

Based on the recording for the period of 9 days, the temperature difference ΔT, the regression coefficient α, and the correlation coefficient r were calculated by the method described above, and the cumulative relative frequency distributions shown in FIGS. 6, 7, and 8 were obtained. Based on these cumulative relative frequency distributions, the reference value Tx of the temperature difference ΔT is set to “1.3 ° C. difference” by the method described above, and the reference value αx of the regression coefficient α is set to “0.4 ° C./10 minutes”. The cooking start time in the 9 day period was estimated. However, in this example, the correlation coefficient r was not used for the cooking start time estimation condition. Table 1 shows the estimated cooking start time. If the estimated cooking start time is within one hour from the previous estimated cooking start time, it was considered as a series of cooking actions and excluded. Moreover, it excluded also when the estimated cooking start time is the same time as the time when the air conditioner 4 stopped.

  In the estimation result of the cooking start time on weekdays, the variation in estimated time by day is not large, such as morning cooking around 7 o'clock, lunch cooking at 12:00 to 13:00, and last cooking after 22:00. Therefore, from the stability of the estimated cooking pattern, it is considered that the cooking start time can be estimated generally well only by the room temperature regression coefficient α and the temperature difference ΔT of the kitchen 1a. In addition, it is thought that it is reheating after a family return home that there is a plurality of cooking start at night.

  FIG. 9 shows the room temperature of each room at 6:45 to 7:45 on the first day. In FIG. 9, the graph indicated by the symbol A indicates the time change of the room temperature in the living room, the graph indicated by the symbol B indicates the time change of the room temperature of the kitchen, the graph indicated by the symbol C indicates the time change of the room temperature of the couple's bedroom, A graph indicated by a symbol D indicates a time change of the room temperature of the child room, and a graph indicated by a symbol E indicates a time change of the room temperature of the Japanese room. From this figure, it can be confirmed that when the room temperature rise of the kitchen 1a is large, the room temperature of the chamber 1b other than the kitchen 1a is also raised. The difference between the maximum temperature and the minimum temperature in the kitchen, living room, couple's bedroom, children's room, and Japanese room in the above hour is 2.3 ° C, 2 ° C, 3 ° C, 2.3 ° C, It is 7 ° C, and the temperature rise of the kitchen 1a is not maximized. This is considered to be due to the difference in the air flow requirement setting for each room and the difference in the positional relationship between the room controller 2 that detects the room temperature and the duct air inlet.

  Under the conditions of Example 1, the reference value Tx of the temperature difference ΔT of the kitchen 1a was relaxed from “1.3 ° C. difference” to “1 ° C. difference”, and the cooking start time was estimated again. As a result, in addition to the times shown in Table 1, three times of 1:34 on the fourth day, 12:13 on the eighth day, and 7:08 on the ninth day were estimated as cooking start times. Referring to the kitchen room temperature on the fourth day shown in FIG. 13, the kitchen room temperature on the eighth day shown in FIG. 17, and the kitchen room temperature on the ninth day shown in FIG. 18, the possibility of starting cooking is high on the seventh day. : Only 08 is considered.

  Here, the average value r_ave of the room temperature correlation coefficient r between the kitchen 1a and the other room 1b is 0.85 at 1:34 on the fourth day, 0.73 at 12:13 on the eighth day, and 9 days. It was 0.95 at 7:08. The cumulative relative frequency when the reference value Tx is “1 ° C. difference” is 90% from FIG. 7. On the other hand, the average value r_ave of the correlation coefficient r corresponding to the cumulative relative frequency of 90% is about 0.9 from FIG. Therefore, the average value r_ave of the correlation coefficient r corresponding to the cumulative relative frequency equivalent to the cumulative relative frequency of the reference value Tx is set as the reference value Rx, and “average value r_ave ≧ reference value Rx” is also added to the cooking start time estimation condition. Thus, when the reference value Tx of the temperature difference ΔT is loosened, it is considered that it is possible to reduce erroneous determination of temperature rise other than cooking. That is, it is considered that the estimation accuracy can be increased by using the correlation coefficient r together.

  The above-described embodiment is an example of a preferred embodiment of the present invention, but is not limited thereto, and various modifications can be made without departing from the gist of the present invention. For example, in the above-described embodiment, the cooking start time is estimated when all three conditions of ΔT ≧ Tx, α ≧ αx, and r_ave ≧ Rx are satisfied, but the processing speed is required rather than the accuracy. In this case, the condition r_ave ≧ Rx may be removed. In addition, when it is desired to capture all the times that are likely to be cooking start times, the cooking start time is estimated when at least one of the conditions of ΔT ≧ Tx and α ≧ αx is satisfied. Anyway.

  Moreover, in the above-described embodiment, the cooking start time is estimated using the target room 1a as a kitchen, but large heat generation similar to cooking, for example, detection of occurrence of a fire or estimation of the occurrence time of a fire, The present invention may be used. In the above-described embodiment, the target room 1a is one room, but each room 1 of the building may be the target room 1a. Further, the cross-correlation may be calculated instead of the correlation coefficient r. By calculating the cross-correlation, information is also obtained as to which chamber 1 has become hot first.

  According to the present invention, the cooking start pattern of the resident can be reflected in the air conditioning by estimating the cooking start time. For example, it is possible to perform air-conditioning that predicts resident behavior, such as starting pre-cooling operation before cooking where heat load increases rapidly, and energy-saving operation such as a whole-building air-conditioning system can be realized. In addition, the estimation result of the cooking start time, in other words, information on whether cooking was performed in a relatively regular time zone is effective as one information for determining whether the resident is living normally. It can be used as a safety confirmation system. Furthermore, according to the present invention, in addition to cooking, a large amount of heat generation, for example, heat generation due to a fire, can be detected, and the generation time can be estimated.

It is a schematic block diagram which shows one Embodiment of the generation | occurrence | production estimation apparatus of the large heat generation of this invention. It is a flowchart which shows an example of a process at the time of applying the generation | occurrence | production estimation method of the large amount of heat generation of this invention to estimation of cooking start time. It is a graph which shows an example of the room temperature information of each room of a building. It is a graph which shows an example of the regression coefficient and temperature difference of kitchen room temperature. An example of the correlation between the room temperature in the kitchen and the room temperature in the other room is shown. (A) shows the correlation with the room temperature in the living room, (B) shows the correlation with the room temperature in the couple's bedroom, and (C) shows the room temperature in the child room. (D) shows the correlation with the room temperature of the Japanese-style room. The cumulative relative frequency distribution of the regression coefficient of kitchen room temperature is shown. The cumulative relative frequency distribution of the temperature difference of kitchen room temperature is shown. The cumulative relative frequency distribution of the room temperature correlation coefficient between the kitchen and other rooms is shown. It is a graph which shows an example of the room temperature information of each room of the building in an Example. It is a graph which shows the room temperature information and fan air volume of the kitchen which were measured in the Example, and shows the data of the 1st day. It is a graph which shows the room temperature information and fan air volume of the kitchen which were measured in the Example, and shows the data of the 2nd day. It is a graph which shows the room temperature information and fan air volume of the kitchen which were measured in the Example, and shows the data of the 3rd day. It is a graph which shows the room temperature information and fan air volume of the kitchen which were measured in the Example, and shows the data of the 4th day. It is a graph which shows the room temperature information and fan air volume of the kitchen which were measured in the Example, and shows the data of the 5th day. It is a graph which shows the room temperature information and fan air volume which were measured in the Example, and shows the data of the 6th day. It is a graph which shows the room temperature information and fan air volume which were measured in the Example, and shows the data of the 7th day. It is a graph which shows the room temperature information and fan air volume which were measured in the Example, and shows the data of the 8th day. It is a graph which shows the room temperature information and fan air volume which were measured in the Example, and shows the data of the 9th day.

Explanation of symbols

1a target room 1b room 3 other than target room 3 temperature sensor 10 information recording means 11 calculation means 12 comparison means

Claims (6)

Record the room temperature information from the temperature sensor installed in the target room and the time information corresponding to the room temperature information, and based on the recorded information, the temperature difference between the maximum room temperature and the minimum room temperature in a predetermined time interval A regression coefficient derived from a time change of room temperature in the time interval is obtained, and this condition is satisfied on condition that one or both of the temperature difference and the regression coefficient are equal to or greater than a predetermined reference value. It is estimated that a larger amount of heat is generated in the target room than in the target room or in a room other than the target room in the time zone corresponding to the time interval , and the reference value of the temperature difference is the temperature difference calculated cumulative frequency line, and wherein the value der Rukoto falls small point variation, the slope of the value or the inclination corresponding to that equal to or less than a predetermined value exceeds a predetermined value of the cumulative frequency line Generation method of estimating the large amount of heat that.   A cumulative frequency line of the regression coefficient is obtained, and a value corresponding to a point where the slope of the cumulative frequency line is equal to or less than a predetermined value or a value corresponding to a point where the slope is smaller than a predetermined value is reduced. The generation estimation method for a large amount of heat generation according to claim 1, wherein the generation value is determined as a reference value for the heat generation. Record the room temperature information from the temperature sensor installed in the target room and the time information corresponding to the room temperature information, and based on the recorded information, the temperature difference between the maximum room temperature and the minimum room temperature in a predetermined time interval A regression coefficient derived from a time change of room temperature in the time interval is obtained, and this condition is satisfied on condition that one or both of the temperature difference and the regression coefficient are equal to or greater than a predetermined reference value. It is estimated that heat generation is greater in the target room than in the target room or in a room other than the target room in the time zone corresponding to the time interval, and the reference value of the regression coefficient is the regression coefficient A cumulative frequency line is obtained, and is a value corresponding to a point at which the slope of the cumulative frequency line is equal to or less than a predetermined value or a value corresponding to a point at which the slope decreases with a change amount exceeding a predetermined value. Generating method of estimating the large amount heat generation and.   The said target room is a kitchen and estimates the generation | occurrence | production estimation method of the large amount of heat generation as described in any one of Claim 1 to 3 which estimates the time estimated that the said big heat_generation | fever occurred as the cooking start time.   The target room is a room of a building equipped with a whole-building air conditioning system, and the condition that the correlation coefficient between the room temperature of the target room and the room temperature of the other room in the time interval is equal to or greater than a predetermined reference value. The method for estimating the occurrence of a large amount of heat according to any one of claims 1 to 4, wherein Based on the information recorded in the information recording means, information recording means for recording the room temperature information from the temperature sensor installed in each room of the building equipped with the entire building air conditioning system and the time information corresponding to the room temperature information, A temperature difference between the highest room temperature and the lowest room temperature of the target room in a predetermined time interval, a regression coefficient derived from a time change of the room temperature of the target room in the time interval, and the room temperature and the other room of the target room in the time interval Calculating means for obtaining a part or all of the correlation coefficient with room temperature, and any or all of the temperature difference, the regression coefficient, and the correlation coefficient are equal to or greater than a predetermined reference value. And a comparison means for determining whether or not this condition is satisfied. The reference value of the temperature difference is a cumulative frequency line of the temperature difference, and a slope of the cumulative frequency line is predetermined. Below value A value corresponding to a point or a value corresponding to a point where the slope becomes smaller with an amount of change exceeding a predetermined value, the reference value of the regression coefficient is a cumulative frequency line of the regression coefficient, and the slope of the cumulative frequency line is A value corresponding to a point that is equal to or less than a predetermined value or a value corresponding to a point where the inclination becomes smaller by an amount of change exceeding a predetermined value, and in the time zone corresponding to the time interval in which the condition is satisfied, the target room A large amount of heat generation estimation apparatus, characterized in that it is estimated that cooking has been performed or a large heat generation similar to cooking has occurred.

Images