JP7272877B2 - Thermal sensation model generation device and method - Google Patents

Description

TECHNICAL FIELD The present invention relates to a technology for generating a thermal sensation model for estimating a resident's thermal sensation in an evaluation target space at a future date and time.

In air-conditioning control and operation management of indoor environments, there is a demand for technology that accurately grasps the degree of satisfaction of residents with indoor environments.
Information on the thermal sensation of the resident is obtained from reports from the residents themselves, their actions, images, physiological measurements, etc., and the obtained thermal sensation information and the index related to the resident's environment at the time of occurrence of the thermal sensation information ( There is a technology that builds a learning model that has learned the relationship with a living environment index), and uses this learned model as an individual thermal sensation model to estimate the thermal sensation of a target resident in an arbitrary environment. The thermal sensation information of the occupant includes a measure of whole-body thermal sensation such as hot/cold, comfort/satisfaction caused by the thermal environment, and the like. Living environment indices include physical environment indices such as temperature and humidity, or PMV (Predicted Mean Vote), which is an index using physical environment indices.

Patent Literature 1 describes a method of substantially learning the relationship between resident thermal sensation information (thermal sensation report in Patent Literature 1) and living environment index (indoor temperature in Patent Literature 1). .
In addition, Non-Patent Document 1 describes a method of learning the relationship between the user's thermal sensation information (discomfort report in Non-Patent Document 1) and the user's living environment index (PMV in Non-Patent Document 1). there is

Residents' thermal sensations constantly change under the influence of seasonally changing factors (hereinafter referred to as "seasonal changing factors") such as seasonal changes in clothing, changes in metabolic rate, and room temperature in the bedroom at home (non-patent literature). 2). For this reason, there is a problem that an estimation error occurs when a trained model based on past data is applied to estimation of thermal sensation information after learning, and an improvement in estimation accuracy is required.

Japanese Patent No. 2713532

Noritoshi Tanaka et al., “Study on ZEB conversion of existing office (4th report) Study of air conditioning control method considering individual thermal sensation”, Proceedings of Annual Meeting of the Society of Heating, Air-Conditioning and Sanitary Engineers of Japan, Vol.10, pp213- 216, 2017/9 Hiroko Kubo et al., "Experimental study on seasonal differences in thermal sensation and comfort sensation", Human and Living Environment, Vol.1, No.1, pp51-57, 1994

SUMMARY OF THE INVENTION It is an object of the present invention to provide a thermal sensation model generating apparatus and method that can bring the estimation result of a resident's thermal sensation closer to the actual situation.

The thermal sensation model generation device of the present invention provides update timing information and learning data period information of a first model that models the relationship between the living environment index of the evaluation target space and the thermal sensation of the resident with respect to the evaluation target space. and the first model based on the living environment index and the resident's thermal sensation information in the learning data period when the update timing comes. and a plurality of first models respectively generated by a plurality of learning data periods and a past outdoor thermal environment corresponding to each of the plurality of learning data periods seasonal dependence management unit configured to obtain a characteristic quantity correction function that indicates the relationship between the model characteristic quantity that determines the shape of the function of the first model and the representative value of the outdoor thermal environment, based on the representative value of Then, the future learning data period to be set for the future date and time for which the resident's thermal sensation is to be estimated is determined, and the model characteristic quantity corresponding to the representative value of the outdoor thermal environment in this learning data period is calculated as described above. a second model generation unit configured to generate a second model obtained by calculating based on the characteristic amount correction function and modifying the latest first model based on the calculated model characteristic amount. It is characterized by

Further, in one configuration example of the thermal sensation model generation device of the present invention, the data period management unit stores information on an overlap period in addition to the information on the update timing and the information on the learning data period. The learning data period of (1) is characterized in that successive learning data periods before and after are set so as to overlap by the width of the overlap period.
Further, one configuration example of the thermal sensation model generation device of the present invention further includes an outdoor thermal environment information management unit configured to store past outdoor thermal environment information of a building containing the evaluation target space, The season-dependent management unit acquires the past outdoor thermal environment information corresponding to each of the plurality of learning data periods from the outdoor thermal environment information management unit, and obtains the past outdoor thermal environment information corresponding to each of the plurality of learning data periods. A representative value of the thermal environment is determined to obtain the characteristic value correction function, and the second model generation unit converts the past outdoor thermal environment information corresponding to the future learning data period into the outdoor thermal environment information management unit. and determining a representative value of the outdoor thermal environment corresponding to the future learning data period.
Further, in one configuration example of the thermal sensation model generation device of the present invention, the second model generation unit includes the model characteristic amount calculated based on the characteristic amount correction function and the latest model characteristic amount of the first model. and correcting the latest first model.
In one configuration example of the thermal sensation model generation device of the present invention, the living environment index is PMV, and the thermal sensation of residents is the proportion of residents who report dissatisfaction with the thermal environment.

In addition, the thermal sensation model generation method of the present invention includes information on the update timing of the first model that models the relationship between the living environment index of the evaluation target space and the thermal sensation of the resident with respect to the evaluation target space, and the learning data period. and referring to the data period management unit that stores the information, and when the update timing comes, the first model is updated based on the living environment index and the resident's thermal sensation information in the learning data period. Based on the first step of generating, a plurality of the first models respectively generated by a plurality of learning data periods, and a representative value of the past outdoor thermal environment corresponding to each of the plurality of learning data periods, a second step of obtaining a characteristic quantity correction function indicating the relationship between the model characteristic quantity that determines the shape of the function of the first model and the representative value of the outdoor thermal environment; A future learning data period to be set for a date and time is determined, and a model characteristic quantity corresponding to a representative value of the outdoor thermal environment in this learning data period is calculated based on the characteristic quantity correction function. and a third step of generating a second model obtained by modifying the latest first model based on the model characteristic quantity.

In one configuration example of the thermal sensation model generation method of the present invention, the data period management unit stores information on an overlap period in addition to the information on the update timing and the information on the learning data period, and stores the information on the overlap period. The learning data period of (1) is characterized in that successive learning data periods before and after are set so as to overlap by the width of the overlap period.
In one configuration example of the thermal sensation model generation method of the present invention, the second step refers to an outdoor thermal environment information management unit that stores past outdoor thermal environment information of a building that includes the evaluation target space. obtaining the past outdoor thermal environment information corresponding to each of the plurality of learning data periods from the outdoor thermal environment information management unit, and obtaining a representative value of the past outdoor thermal environment corresponding to each of the plurality of learning data periods; and obtaining the characteristic quantity correction function, wherein the third step acquires the past outdoor thermal environment information corresponding to the future learning data period from the outdoor thermal environment information management unit. determining a representative value of the outdoor thermal environment corresponding to the future learning data period.
Further, in one configuration example of the thermal sensation model generation method of the present invention, the third step includes adding the model characteristic amount calculated based on the characteristic amount correction function and the latest model characteristic amount of the first model. and modifying the latest first model based on.

According to the present invention, a data period management unit, a first model generation unit, a season dependence management unit, and a second model generation unit are provided, and the first model is generated based on the outdoor thermal environment state reflecting the seasonal transition. can be modified to generate the second model, the estimation error due to the seasonal transition factor can be improved. As a result, in the present invention, it is possible to bring the estimation result of the resident's thermal sensation closer to the actual situation.

Further, in the present invention, in addition to the update timing information and learning data period information, overlap period information is stored in the data period management unit, and a plurality of consecutive learning data periods are stored by the width of the overlap period. By setting them to overlap, the reliability of the model can be further improved.

FIG. 1 is a diagram showing an example of an air conditioning control system according to a first embodiment of the invention. FIG. 2 is a block diagram showing the configuration of the thermal sensation model generation device according to the first embodiment of the present invention. FIG. 3 is a flow chart for explaining the operations of the living environment information holding unit, the thermal sensation information holding unit, and the data integration unit of the thermal sensation model generating apparatus according to the first embodiment of the present invention. FIG. 4 is a diagram showing an example of thermal sensation report information, report date and time, and reporter ID in the first embodiment of the present invention. FIG. 5 is a diagram showing an example of integrated information in the first embodiment of the invention. FIG. 6 is a flow chart for explaining the operation of the past model generation unit of the thermal sensation model generation device according to the first embodiment of the present invention. FIG. 7 is a diagram showing the relationship between PMV, which is a living environment index, and the number of people reporting "hot". FIG. 8 is a flow chart explaining a method of generating modeling information in the first embodiment of the present invention. FIG. 9 is a diagram showing the relationship between PMV and PPD. FIG. 10 is a diagram showing an example of a dissatisfaction level function. FIG. 11 is a flow chart for explaining the operation of the seasonal dependence management section of the thermal sensation model generating apparatus according to the first embodiment of the present invention. FIG. 12 is a flow chart for explaining the characteristic quantity correction function determination operation of the seasonal dependence management unit of the thermal sensation model generating apparatus according to the first embodiment of the present invention. FIG. 13 is a diagram for explaining model characteristic quantities in the first embodiment of the present invention. 14A and 14B are diagrams for explaining the method of determining the characteristic quantity correction function in the first embodiment of the present invention. FIG. FIG. 15 is a diagram explaining another model characteristic amount in the first embodiment of the present invention. FIG. 16 is a flow chart for explaining the operation of the future model generation unit of the thermal sensation model generation device according to the first embodiment of the present invention. FIG. 17 is a diagram for explaining learning data periods in the first and second embodiments of the present invention. FIG. 18 is a diagram showing an example of the period average transition value of the outside air temperature. FIG. 19 is a block diagram showing a configuration example of a computer that implements the thermal sensation model generation device according to the first and second embodiments of the present invention.

[Principle of Invention]
If there is a seasonal transition, it is necessary to deal with the fact that the actual state of thermal sensation has changed at the time when the data for learning was collected. Therefore, the inventor focused on correcting an individual thermal sensation model (hereinafter referred to as a past model) that has been learned based on past data based on the outdoor thermal environment state in which the seasonal transition is reflected, and making it a future model. By correcting the past model with the outdoor thermal environment conditions that reflect seasonal changes, it is possible to reflect the effects of seasonal changes in the model.

Outdoor thermal environment conditions that reflect seasonal changes include the outdoor air temperature. For example, a future model can be generated from a trained past model by using the average outside air temperature during the learning data period when generating a past model, and the future average outside temperature and forecast outside temperature for which you want to estimate the thermal perception of residents. do.

Furthermore, the inventor noted that the reliability of the past model and the future model cannot be maintained when using data for a period of two days, three days, or the like that includes short-term fluctuations. In order to take into consideration seasonal changes, which are annual fluctuations, it is necessary to secure data with a certain period width corresponding to the period of this fluctuation for model learning. A period of about one month is suitable for this period, but in order to maintain the continuity of the seasonal transition, an overlap method is adopted in which a reliable period (such as two weeks) overlaps the previous period and the latter period. is effective.

[First embodiment]
Hereinafter, embodiments of the present invention will be described with reference to the drawings. A first embodiment of the present invention describes a basic technique for generating a future model from a past model.
This embodiment is an example of generating a model for each month on a monthly basis. is corrected by the characteristic amount correction function, the past model is corrected using the characteristic amount correction function, and the future model is obtained.

In the present embodiment, for the sake of simplicity of explanation, as shown in FIG. 1, an example of a resident-reported air-conditioning control system in which two residents are seated in one target air-conditioning zone will be explained. Regarding the resident reporting type air conditioning control system, for example, see the document "Yasuhito Omagari et al., ``Demonstration test of air conditioning system that responds to thermal sensation reporting,'' The Society of Air-Conditioning and Sanitary Engineers of Japan, Vol. 3, pp. 41-44, 2016. /9".

In FIG. 1, 100 is a resident, 101 is an evaluation target space (a space where the resident is seated), 102 is an air conditioning control device (controller) that receives thermal sensation reports, and 103 is the temperature for measuring the room temperature of the evaluation target space 101. Sensors, 104 is a humidity sensor for measuring the humidity of the evaluation target space 101, 105 is an indoor unit, and 106 is an outdoor unit. The air conditioning control device 102 controls the air conditioning equipment (indoor unit 105 and outdoor unit 106) so that the room temperature measured by the temperature sensor 103 matches the room temperature set value and the humidity measured by the humidity sensor 104 matches the humidity set value. to control.

FIG. 2 is a block diagram showing the configuration of the thermal sensation model generation device of this embodiment. The thermal sensation model generation device is composed of a past model generation unit 1 and a future model generation unit 2 .

The past model generation unit 1 provides the data collection unit 10 with information and learning about the update timing of the individual thermal sensation model that models the relationship between the living environment index of the evaluation target space 101 and the thermal sensation of the occupants of the evaluation target space 101. A data period management unit 14 that stores data period information, and a past model (first model) based on the living environment index and the resident's thermal sensation information in the learning data period when the update timing comes. It includes a past model generation unit 15 (first model generation unit) that generates a past model, and a past model storage management unit 16 that stores the past model generated by the past model generation unit 15 .

The data collection unit 10 includes a living environment information storage unit 11 that calculates a living environment index from the thermal environment information, a thermal sensation information storage unit 12 that acquires the thermal sensation information reported by the residents of the evaluation target space, and a thermal sensation information storage unit 12 that acquires and a data integration unit 13 that extracts the living environment index calculated from the thermal environment information of the measurement date and time corresponding to the date and time of occurrence of the information, and generates integrated information that integrates the thermal sensation information and the living environment index. .

The future model generation unit 2 includes an outdoor thermal environment information management unit 20 that stores past outdoor thermal environment information of the building that includes the evaluation target space, and a plurality of past models and a plurality of past models respectively generated by a plurality of learning data periods. Based on the representative value of the past outdoor thermal environment corresponding to each learning data period, a characteristic value correction function that indicates the relationship between the model characteristic value that determines the shape of the function of the past model and the representative value of the outdoor thermal environment. Determine the future learning data period to be set for the desired season-dependent management unit 21 and the future date and time for estimating the resident's thermal sensation, and correspond to the representative value of the outdoor thermal environment in this learning data period. A future model generation unit 22 (second model generator).

FIG. 3 is a flow chart for explaining the operation of the living environment information storage unit 11, the thermal sensation information storage unit 12, and the data integration unit 13. As shown in FIG.
The living environment information holding unit 11 acquires thermal environment information (thermal environment measurement values such as temperature measurement values and humidity measurement values) from, for example, environment measurement devices (temperature sensor 103, humidity sensor 104, etc.) installed in the evaluation target space 101. Receive and calculate the living environment index. Then, the living environment information holding unit 11 associates the measurement date and time of the thermal environment measurement value with the living environment index and stores them as living environment information (measurement date and time, living environment index) (step S100 in FIG. 3).

The living environment information holding unit 11 stores information such as calculation formulas and parameter values necessary for calculating living environment indices, types of thermal environment measurement values used for calculation, and other information provided by solution providers such as indoor environment evaluation/control and equipment management. preset by the user.

The living environment information holding unit 11 transmits the living environment information (measurement date and time, living environment index) to the data integrating unit 13 in response to the information transmission request from the data integrating unit 13 (step S101 in FIG. 3).

The living environment index is a thermal environmental index of the living environment, and is a physical environment index such as air temperature and humidity that are generally measured in an air-conditioned environment, or a general index calculated using a physical environment index (e.g. working temperature, SET ^* (Standard new Effective Temperature), PMV, etc.). In this embodiment, PMV, which is internationally standardized, is used as the living environment index, and six elements normally required for calculation of PMV: air temperature Ta [° C.], humidity H [%], average radiation temperature Tr [° C.], airflow Velocity v [m/s], metabolic rate M [met], and clothing amount Icl [clo] are set to the following values. That is, the air temperature Ta [°C] and humidity H [%] are the air temperature and humidity measurement values of the target building, the average radiation temperature Tr is equal to the air temperature Ta, and the air velocity v is 0.1 [m/s], the metabolic rate M is 1.0 [met], and the amount of clothing Icl is 0.8 [clo]. It is assumed that other information necessary for PMV calculation is set with reference to international standards and the like together with PMV calculation formulas.

Here, a method for calculating PMV will be described. Since the calculation method of PMV is internationally standardized by ISO-7730, the calculation method of this standard may be followed. PMV is obtained using the following formulas (1) to (3). The calculation program is also published in ANSI/ASHRAE Standard 55-2010 and the like.
PMV=Q(M)×L (1)
Q(M)=0.303exp ^-0.036M +0.028 (2)
L=M-(C+R+Ed+Es)-(Cre+Ere) (3)

M is the metabolic rate [met], L is the heat load of the human body L [W/m ² ], C is the convective heat loss [W/m ² ], R is the radiation heat loss [W/m ² ], and Ed is the The amount of insensible heat [W/m ² ], Es is the amount of evaporative heat loss due to perspiration [W/m ² ], Cre is the amount of sensible heat loss due to respiration [W/m ² ], Ere is the amount of latent heat loss due to respiration [W /m ² ].

Convergence calculation of the following equations (4) to (6) is performed to calculate the amount of convective heat loss C and the amount of radiation heat loss R. tcl is the clothing outer surface temperature [°C].
C=fcl×hc×(tcl−Ta) (4)
R=3.96×10 ⁻⁸ ×fcl×{(tcl+273.15) ⁴
−(Tr+273.15) ⁴ } (5)
tcl=ts−0.155×Icl×(C+R) (6)

In equations (4) to (6), Icl is the amount of clothing [clo]. fcl is a clothing area increase coefficient, fcl=1+1.29×Icl when Icl<0.78, and fcl=1.05+0.645×Icl when Icl≧0.78. Tr is the average radiation temperature [°C]. ts is the average skin temperature [°C] and is obtained by the following equation. W is mechanical work [W/m ² ] (usually 0).
ts=35.7−0.028×(M−W) (7)

Also, hc is the convective heat transfer coefficient of the human body [W/(m ² ·°C)] and is the larger of 2.38 x |tcl-ta| ^0.25 or 12.1 x v ^0.5 . v is the air velocity [m/s].

The dead heat amount Ed is obtained by the following equation.
Ed=3.05×(5.73−0.007×Mpa) (8)
The evaporative heat loss amount Es due to perspiration is obtained by the following equation.
Es=0.42×(M−58.15) (9)

The amount of sensible heat loss Cre due to respiration is obtained by the following equation.
Cre=0.0014×M×(34−Ta) (10)
The latent heat loss amount Ere due to respiration is obtained by the following equation. pa is water vapor pressure.
Ere=0.0173×M×(5.87-pa) (11)

As described above, the living environment information holding unit 11 can calculate the PMV as the living environment index. In the present embodiment, it is assumed that the thermal environment measurement value measured in a 1-minute cycle is transmitted to the living environment information holding unit 11, and the living environment information holding unit 11 calculates the PMV in a 1-minute cycle. If thermal environment measurement values are periodically stored in an air conditioning control device or a central monitoring system that manages a building, the thermal environment measurement values can be acquired from the air conditioning control device or central monitoring system.

On the other hand, the thermal sensation information holding unit 12 receives the thermal sensation information of the occupant of the evaluation target space 101, associates the occurrence date and time of the thermal sensation information with the information of the resident who declared the thermal sensation information, and stores the thermal sensation information. Store (step S102 in FIG. 3).

The thermal sensation information holding unit 12 transmits the thermal sensation information, the occurrence date and time information, and the resident information to the data integration unit 13 in response to the information transmission request from the data integration unit 13 (step S103 in FIG. 3).

The thermal sensation information is a quantity that indicates how the resident feels in relation to the living environment index, and indicates information such as a measure of whole-body thermal sensation such as hot/cold, and comfort/satisfaction caused by the thermal environment. . Indices such as the number of dissatisfied persons in the target space and the rate of dissatisfied persons calculated from the number of dissatisfied persons are also included in the thermal sensation information. A smart phone, a PC (personal computer), or the like may be appropriately used as an input terminal for the thermal sensation information.

In this embodiment, thermal sensation report information (hot/cold) from the resident is used as the thermal sensation information, and the resident freely declares hot or cold thermal sensation through the input terminal. FIG. 4 shows an example of thermal sensation report information, report date and time (occurrence date and time information), and reporter ID (resident information). In the example of FIG. 4, "hot" indicates hot and "cold" indicates cold.

The reporter ID may be input by the resident together with the thermal sensation report and sent to the thermal sensation information holding unit 12, or when each resident uses an input terminal, the terminal ID is used as the reporter ID. , may be transmitted to the thermal sensation information storage unit 12 .
The reported date and time may be transmitted from the input terminal to the thermal sensation information storage unit 12, or the date and time when the thermal sensation information storage unit 12 receives the thermal sensation report information may be used as the reported date and time.

Next, when the data integration unit 13 receives the living environment index from the living environment information holding unit 11 and the thermal sensation information from the thermal sensation information holding unit 12, the data integration unit 13 receives the living environment index at the measurement date and time corresponding to the occurrence date and time of the thermal sensation information. The environmental index is extracted, and integrated information (occurrence date/time information, resident ID, thermal sensation information, living environment index) obtained by integrating the thermal sensation information and the living environment index is stored (step S104 in FIG. 3).

The data integration unit 13 transmits integrated information to the past model generation unit 15 in response to the information transmission request from the past model generation unit 15 (step S105 in FIG. 3).

FIG. 5 shows an example of the integrated information including the heat sensation information generation date and time information (reported date and time information), the resident ID (reporter ID), the heat sensation information, and the living environment index (PMV).
When extracting the living environment index of the measurement date and time corresponding to the date and time of occurrence of the thermal sensation information, the resolution of the date and time of occurrence of the thermal sensation information (reported date and time) is, for example, 1 minute, and the resolution of the measurement date and time of the thermal environment measurement value is For example, the time interval between the generation of the thermal sensation information and the measurement of the thermal environment measured value may be different, such as every 10 minutes. In this case, the solution provider or facility manager should determine the extraction rule in advance.

For example, if the resolution of the actual thermal information occurrence date and time (declared date and time) is in units of one minute, and the thermal environment measurement value is measured every 10 minutes, the measurement date and time closest to the actual thermal information occurrence date and time, or An extraction rule may be determined such that the date and time of occurrence advanced by 10 minutes is used as the measurement date and time corresponding to the date and time of occurrence of the thermal sensation information.

When the thermal environment measurement value is measured every 10 minutes such as 14:50, 15:00, 15:10, etc., the measurement corresponding to the date and time of occurrence of the thermal sensation information at 15:03 The date and time is 15:00, or 15:10, which is the latest 15:00 or 15:10, which is advanced by 10 minutes from 15:03.

The living environment information storage unit 11, the thermal sensation information storage unit 12, and the data integration unit 13 continuously execute the processing of FIG. 3, and the data integration unit 13 accumulates integrated information.

Next, the data period management unit 14 stores data period management information for determining the update timing of the individual thermal sensation model (past model) and the learning data period. preset by.
The data period management unit 14 transmits data period management information in response to an information transmission request from the past model generation unit 15 or the future model generation unit 22 .

In this embodiment, the information on the update timing of the individual thermal sensation model and the information on the learning data period width are used as the data period management information. Specifically, the update timing of the individual thermal sensation model is set to 0:00 on the 1st day of every month (X month 1, X is an arbitrary month), and the learning data period range is changed from the previous update timing to the current update. One month until the timing. In other words, since the individual thermal sensation model is updated on the first day of every month, the start date and time of the learning data period for updating the individual thermal sensation model is 0:00 on the first day of the previous month, and the end date and time of the learning data period. is 23:59 on the last day of the previous month.

Since it is sufficient to determine the update timing and the learning data period of the individual thermal sensation model from the data period management information, the update start date and time and the learning data period range (every 14 days, 168 hours, etc.) going back from the update start date and time may be set as the data period management information.

FIG. 6 is a flowchart for explaining the operation of the past model generator 15. As shown in FIG. Based on the data period management information acquired from the data period management unit 14, the past model generation unit 15 recognizes that it is time to update the individual thermal sensation model (YES in step S200 in FIG. 6), and responds to the current model update. The start date and time and the end date and time of the learning data period to be used are determined (step S201 in FIG. 6).

The past model generation unit 15 acquires the integrated information of the determined learning data period from the data integration unit 13 (step S202 in FIG. 6), and uses the acquired integrated information to input/output data for identifying the individual thermal sensation model. Data (modeling information) is generated (step S203 in FIG. 6). The past model generator 15 uses the generated modeling information to identify the individual thermal sensation model, and updates this individual thermal sensation model as the latest past model (step S204 in FIG. 6).

Then, the past model generation unit 15 transmits the modeling result of the latest past model and information on the learning data period used for modeling to the past model storage management unit 16 (step S205 in FIG. 6).
The past model generation unit 15 performs the processing of FIG. 6 at each model update timing.

In this embodiment, as described above, the individual thermal sensation model is updated at 0:00 on the first day of every month. For example, when the individual thermal sensation model is updated at 0:00 on March 1, 2018, the learning data period is from 0:00 on February 1, 2018 to 23:59 on February 28, 2018. When the individual thermal sensation model is updated at 0:00 on August 1, 2019, the learning data period is from 0:00 on July 1, 2018 to 23:59 on July 31, 2018.

In addition, in the present embodiment, the living environment index (PMV) and the report rate of "hot" PPVh are used as modeling information, and the function of the predicted dissatisfied rate PPD (Predicted Percentage Dissatisfied) is used as the basic function of the individual thermal sensation model. do. PPD is also internationally standardized, and ISO-7730 shows an arithmetic expression using PMV.

Next, a method of generating modeling information in step S203 will be described. As described above, the past model generator 15 generates modeling information from integrated information.
The report rate PPVh of "It's hot" in any PMV is calculated by the following equation, where N is the number of people reporting "It's hot" in any PMV, and N_all is the number of residents in the evaluation target space.
PPVh=N/N_all (12)

The number of residents N_all in the space to be evaluated is set in advance by a solution provider such as indoor environment evaluation/control or a facility manager.
Here, the number of reporters N in Equation (12) may be the number of reporters calculated in consideration of the physiological validity of the thermal sensation. This calculation method is based on the principle that a person who declares "hot" in an arbitrary PMV feels hotter in a PMV on the hotter side than the environment.

Specifically, each reporter sets the first reported PMV when the PMV is low (weak heat) to the latch start point PMVstart(m) (m is a positive integer, m= 1 to 4), and in the PMV area larger than the latch start point PMVstart(m), the number of people is calculated while maintaining the occurrence state of the report "hot". According to this calculation method, as shown in FIG. 7, the number of people reporting "hot" increases as the environment becomes hotter, and the physiological validity becomes higher. Hereinafter, this method will be referred to as the declaration latch method, and an example of generating modeling information by this declaration latch method will be described.

FIG. 8 shows a flow chart for calculating the report rate PPVh from the number of reporters calculated using the report latch method and generating modeling information. The past model generation unit 15 extracts only the information of the report "hot" from the integrated information acquired in step S202 (step S300 in FIG. 8), and the number of reporters with different reporter IDs included in the extracted information (" N_vote (the number of residents who declared that it is "hot") is obtained (step S301 in FIG. 8). Assuming that the data example shown in FIG. 5 is integrated information for the learning data period, the number of voters who voted "hot" is N_vote=4.

Next, the past model generator 15 initializes the variable m (m is a positive integer equal to or smaller than N_vote) to 1 for counting voters (step S302 in FIG. 8).
The past model generation unit 15 obtains the minimum value of the PMVs when the m-th reporter declares "hot" as the latch start point PMVstart(m) (step S303 in FIG. 8).

The past model generation unit 15 determines whether or not m=N_vote, that is, all applicants who declared “hot” have completed the processing of step S303 (step S304 in FIG. 8), and if the variable m is less than N_vote The variable m is incremented by 1 assuming that the process has not been completed (step S305 in FIG. 8). In this way, the process of step S303 is performed for each reporter who declares "hot".

When m=N_vote, and the process of step S303 has been completed for all applicants who declared that they are “hot” (YES in step S304), the past model generation unit 15 generates an arbitrary PMV and the corresponding “hot” vote. Modeling information consisting of the filer rate PPVh is generated (step S306 in FIG. 8).

The state V_latch(m) of whether or not the filer P(m) reports any PMV is V_latch(m)=0 (no report) when PMV<PMVstart(m), and when PMVstart(m)≦PMV , V_latch(m)=1 (notified).
At this time, the report rate PPVh of "hot" for any given PMV is obtained by the following equation.

Thus, the calculation of the "hot" reporter rate PPVh is completed, and the generation of modeling information including the learning data period PMV and the "hot" reporter rate PPVh in the learning data period is completed. In the case of the data example of FIG. 5, the modeling information is obtained by dividing the vertical axis in FIG. 7 by the number of persons reporting "hot" to obtain the reporting rate of persons reporting "hot". The modeling information may be PMV generated at regular intervals, or may be data only for points where PPVh increases.

Next, a method for identifying the individual thermal sensation model in step S204 will be described. As described above, in this embodiment, the predicted dissatisfied rate PPD is used as the basic function of the individual thermal sensation model. PPD is represented by the following relational expression as a function of PMV.
PPD=F(PMV)=100-95*exp(-0.03353* ^PMV4
−0.2179×PMV ² ) [%] (14)

FIG. 9 shows the relationship between PMV and PPD. The individual thermal sensation model may be defined as a V-shaped distribution function as shown in FIG. 9 in which PMV and PPD are associated with each other. Alternatively, it is possible to separate the dissatisfaction levels of the hot side and the cold side and define a dissatisfaction function (individual thermal sensation model).

FIG. 10 is a diagram showing an example of the dissatisfaction degree function disclosed in JP-A-2015-4480. To obtain the dissatisfaction level of "hot", the dissatisfaction level function indicated by 110 in FIG. 10 can be used, and to obtain the dissatisfaction level of "cold", the dissatisfaction level function indicated by 111 in FIG. 10 is used. do it. A function such as that shown in FIG. 10 is, for example, a sigmoid function, and the function of the individual thermal sensation model may be such a general model.

In this embodiment, the model of the following formula (15), which is the basic function of the formula (14) and is modified by reflecting the resident's thermal feeling, is used as an individual thermal feeling model (dissatisfaction model). do. The shape of the individual thermal sensation model in Equation (15) is similar to the function indicated by 110 in FIG. In equation (15), the predicted dissatisfied person rate PPD in equation (14), which corresponds to the degree of dissatisfaction, is used as the "hot" reporting rate PPVh that reflects the resident's thermal sensation.

In Equation (15), a, b, c, d, and e are parameters of the model to be searched. Formula (16) is obtained by setting a = b (the minimum value of the report rate PPVh of "hot" is 0) in formula (15). The parameters of the model are a, c, d, and e.

The model function of the individual thermal sensation model such as formula (15) or formula (16), the parameters of the model to be searched, and the optimization method to be used are determined by solution providers and facility managers such as indoor environment evaluation / control. , is preset.

The past model generating unit 15 uses the modeling information generated in step S203 to match the model function of the individual thermal sensation model preset by the solution provider or facility manager with the similarly preset search target model. The parameters are explored to identify individual thermal perception models. General-purpose methods such as the least-squares method and the simplex method may be used as the optimization method used when searching for parameters.
Updating of the past model is thus completed.

The past model storage management unit 16 stores the past model modeling results (parameter values of the searched model) received from the past model generation unit 15 and the learning data period information. In response to the transmission request, the stored information is transmitted to the season-dependent management section 21 .

The outdoor thermal environment information management unit 20 stores past outdoor thermal environment information and future outdoor thermal environment information of the building including the evaluation target space 101 . The past outdoor thermal environment information consists of outdoor thermal environment state information and date and time information when the outdoor thermal environment state information was acquired. The future outdoor thermal environment information is information necessary for estimating the future outdoor thermal environment, and consists of outdoor thermal environment state information and date and time information when the appearance of the outdoor thermal environment state information is estimated.

The outdoor thermal environment state information that reflects seasonal changes includes indices such as outdoor temperature and humidity, a discomfort index that integrates temperature and humidity, and WBGT (wet-bulb globe temperature). The past outdoor thermal environment information may be statistics such as normal year data (monthly average temperature, daily average temperature, etc.), or may be measured values obtained by a temperature sensor or the like installed outdoors. The future outdoor thermal environment information is essentially the past outdoor thermal environment information corresponding to the future learning data period. Alternatively, it may be an estimated value of outdoor thermal environment information that is estimated based on a weather forecast or the like.

The method of acquiring the outdoor thermal environment information is set in advance by solution providers such as indoor environment evaluation/control and facility managers. When using data that can be obtained in advance, such as the monthly average temperature, as the outdoor thermal environment state information, these data may be set in the outdoor thermal environment information management unit 20 in advance.
In this embodiment, the monthly average outside air temperature for each month of a normal year is used as past and future outdoor thermal environment state information.

Next, the seasonal dependence management unit 21 determines the characteristic amount correction function Gpm in response to a request from the future model generation unit 22, and compares this characteristic amount correction function Gpm with the past model stored in the past model storage management unit 16. Modeling results (parameter values of the searched model) are sent to the future model generation unit 22 .

FIG. 11 is a flow chart for explaining the operation of the seasonal dependence manager 21. FIG. The seasonal dependence management unit 21 uses N_ref (N_ref is the preset number of reference models) modeling results from the latest modeling result to (N_ref-1) previous modeling results, and these N_ref modeling results. Information on the learning data period is obtained from the past model storage management unit 16 (step S400 in FIG. 11).

Subsequently, the season-dependent manager 21 acquires past outdoor thermal environment information corresponding to each of the learning data periods used for N_ref modeling from the outdoor thermal environment information manager 20 (step S401 in FIG. 11).

Then, based on the obtained past outdoor thermal environment information, the season-dependent management unit 21 calculates, for each learning data period, the past outdoor thermal environment representative value Eot corresponding to the learning data period used for modeling N_ref times. (Step S402 in FIG. 11). For example, the outdoor thermal environment state information acquired from the outdoor thermal environment information management unit 20 in the past is the outside air temperature measured every hour, and the learning data period is from 00:00 on August 1, 2018 to August 10, 2018. If it is 23:59, the average value of outside air temperature during this learning data period (average outside temperature) may be calculated and used as the past outdoor thermal environment representative value Eot.

However, as described above, when using the monthly average outside temperature for each month of a normal year as the past outdoor thermal environment status information, the season-dependent management unit 21 uses the past average outside temperature for the month including the learning data period as it is. outdoor thermal environment representative value Eot. In addition, when the learning data period is set over a plurality of months, if the average value of the monthly average outside temperature of these months or the weighted average value according to the number of days is used as the past outdoor thermal environment representative value Eot good.

Next, the season-dependent management unit 21 obtains a characteristic value correction function Gpm that indicates the relationship between the past outdoor thermal environment representative value Eot and the model characteristic value SZ of the past model (step S403 in FIG. 11). The model characteristic quantity SZ is one of the quantities that determine the shape of the function of the past model. Once the characteristic quantity correction function Gpm is determined, it becomes possible to estimate the model characteristic quantity SZ corresponding to an arbitrary outdoor thermal environment representative value Eot.

FIG. 12 is a flow chart for explaining the operation of determining the characteristic quantity correction function of the seasonal dependence management unit 21. As shown in FIG. In the following description, i is a variable for counting past models and is an integer from 1 to N_ref. A smaller variable i indicates a new past model.

Further, as shown in FIG. 13, the model characteristic quantity SZ is set to the saturated value of the report rate PPVh of "hot", and the characteristic quantity correction function Gpm is the relationship between the past outdoor thermal environment representative value Eot and the model characteristic quantity SZ. Let it be a function of linear approximation. In other words, once an arbitrary outdoor thermal environment representative value Eot is determined, an estimated value of the corresponding model characteristic quantity SZ can be calculated by the characteristic quantity correction function Gpm. When the individual thermal sensation model is set to Equations (15) and (16), the saturation value of the rate of reporters PPVh corresponds to a of the search parameters a, b, c, d, and e of the model.

In step S400, the seasonal dependence management unit 21 obtains N_ref modeling results Sp(i), that is, a set of search parameter values searched for the i-th past model (in this embodiment, the searched parameter a( i), b(i), c(i), d(i), e(i)) and the information of the learning data period LP(i) used for modeling these N_ref times are past model memory management After obtaining from the unit 16, the variable i is initialized to 1 (step S500 in FIG. 12).

The season-dependent management unit 21 acquires the past outdoor thermal environment representative value Eot(i) corresponding to the learning data period LP(i) used for the i-th modeling from the latest from the values calculated in step S402 ( FIG. 12 step S501).

The seasonal dependence management unit 21 determines whether the processing of step S501 has been completed for i=N_ref, that is, all N_ref learning data periods LP(i) (step S502 in FIG. 12), and if the variable i is less than N_ref 1, the variable i is incremented by 1 (step S503 in FIG. 12). Thus, the process of step S501 is performed for each learning data period LP(i).

When i=N_ref and the processing of step S501 is completed for all N_ref learning data periods LP(i) (YES in step S502), the seasonal dependence management unit 21 searches for functions and parameters of the individual thermal sensation model. A function that linearly approximates the relationship between the model characteristic quantity SZ(i) determined from the modeling result Sp(i), which is a set of values, and the outdoor thermal environment representative value Eot(i) is obtained, and this function is called the characteristic quantity correction function. Gpm (step S504 in FIG. 12). Here, in the present embodiment, where the model characteristic quantity SZ is the saturated value of the report rate PPVh of "hot", SZ(i)=a(i).

14(A) and 14(B) are diagrams for explaining a method of determining the characteristic quantity correction function Gpm. In the example of FIGS. 14A and 14B, N_ref=3, the function of the latest past model is Mp(1), the function of the past model one time before the latest is Mp(2), and the Mp(3) is the function of the past model two times before.

For example, the learning data period LP(1) used when modeling the latest past model is September, the learning data period LP(2) used when modeling the previous past model is August, and two times before. If the learning data period LP(3) used for modeling the past model is July, the past outdoor thermal environment representative values Eot(1), Eot(2), and Eot(3) are September , August and July are normal outside temperatures. The model characteristic quantities SZ(1), SZ(2), SZ(3) are the saturation values (parameters a(1), a(2 ), a(3)).

Then, as shown in FIG. 14B, past outdoor thermal environment representative values {Eot(1), Eot(2), Eot(3)} and model characteristic quantities {SZ(1), SZ(2), A function obtained by linearly approximating the relationship with SZ(3)} is defined as a characteristic quantity correction function Gpm. The characteristic quantity correction function Gpm is a function for estimating a model characteristic quantity SZ (saturated value of the report rate PPVh of "hot") corresponding to an arbitrary monthly average outdoor temperature.

In the present embodiment, linear approximation was used as a method for obtaining the characteristic amount correction function Gpm, but the characteristic amount correction function Gpm is obtained by polynomial approximation, exponential approximation, or the like, taking into consideration the influence of seasonal changes on thermal perception. can be determined as appropriate.

In the present embodiment, the saturation value of the report rate PPVh of "hot" is used as the model characteristic quantity SZ to determine the characteristic quantity correction function Gpm. A quantity may be used, and a plurality of types of model characteristic quantities SZ may be set.

For example, the saturated value of the declaring rate PPVh of "hot" is the model characteristic quantity SZ1 (SZ in FIG. 13), the PMV value of the rising position of the declaring rate PPVh as shown in FIG. As shown in 15(B), let the PMV value when the reportr rate PPVh reaches 50% of the saturation value be the model characteristic quantity SZ3.

In the example of FIG. 15A, the model characteristic quantity SZ2 is set to e of the model search parameters a, b, c, d, and e when the individual thermal sensation model is represented by the equations (15) and (16). handle.
Therefore, the seasonal dependence management unit 21 acquires the model characteristic quantity SZ2(i)=e(i) from each of the N_ref modeling results Sp(i), and similarly to the model characteristic quantity SZ1, the model characteristic quantity A function linearly approximating the relationship between SZ2(i)=e(i) and the past outdoor thermal environment representative value Eot(i) may be obtained, and this function may be used as the characteristic quantity correction function Gpm_2 (step S504).

In the example of FIG. 15B, the model characteristic value SZ3 is the PMV value when the reportr rate PPVh reaches 50% of the saturation value. Therefore, the season-dependent management unit 21 calculates the value a_50(i), which is 50% of a(i), from each of the model functions of the individual thermal sensation model and the N_ref modeling results Sp(i). The PMV value that makes the report rate PPVh of a_50(i) is calculated from the formula (16) (or formula (15)) and the modeling result Sp(i), and the calculated result is SZ3(i). Just do it.

Then, as with the model characteristic value SZ1, the seasonal dependence management unit 21 obtains a function that linearly approximates the relationship between the model characteristic value SZ3(i) and the past outdoor thermal environment representative value Eot(i), This function may be used as the characteristic quantity correction function Gpm_3 (step S504). Let Gpm_1 be the characteristic quantity correction function obtained from the model characteristic quantity SZ1 (SZ in FIG. 13).
In this way, a characteristic quantity correction function can be obtained for each type of model characteristic quantity SZ.

Finally, in response to an information transmission request from the future model generation unit 22, the seasonal dependence management unit 21 transmits the characteristic amount correction functions Gpm (Gpm_1, Gpm_2, Gpm_3) and the latest past model modeling results (Fig. 11 step S404).

The number of reference models N_ref, the method of determining the past outdoor thermal environment representative value Eot, the method of determining the model characteristic value SZ, and the method of determining the characteristic value correction function Gpm corresponding to each characteristic value are the indoor environment evaluation/control, etc. preconfigured by your solution provider or facility manager.

FIG. 16 is a flow chart for explaining the operation of the future model generation unit 22. As shown in FIG. The future model generating unit 22 estimates the thermal sensation of the residents of the evaluation target space 101 (the percentage of residents who report dissatisfaction with the thermal environment of the evaluation target space 101, and the reporting rate PPVh of “hot”). date and time (estimated future date and time Tm) is acquired by an instruction from the operator (step S600 in FIG. 16).

The future model generation unit 22 acquires information on the update timing of the individual thermal sensation model from the data period management unit 14, and determines the future learning data period LPF including the estimated future date and time Tm (step S601 in FIG. 16). For example, if the estimated future date and time Tm is October 15, 2019, the update timing of the individual thermal sensation model is 0:00 on the first day of every month, and the learning data period width is one month. determines the future training data period LPF to be October 2019.

Then, the future model generation unit 22 acquires the future outdoor thermal environment information corresponding to the future learning data period LPF from the outdoor thermal environment information management unit 20 (step S602 in FIG. 16), and the acquired future outdoor thermal environment Based on the information, a future outdoor thermal environment representative value Eotf corresponding to the future learning data period LPF is determined (step S603 in FIG. 16). For example, if the future learning data period LPF is October 2019, the future model generation unit 22 may obtain the average monthly average outside air temperature for October as the future outdoor thermal environment representative value Eotf.

Next, the future model generation unit 22 uses the characteristic amount correction function Gpm (here, Gpm_1) transmitted from the season-dependent management unit 21 to calculate the model characteristic amount SZ1 corresponding to the future outdoor thermal environment representative value Eotf. Gpm_1 (Eoft) is calculated, and the calculated value is set as the estimated characteristic value SZest (step S604 in FIG. 16).

Finally, the future model generator 22 generates an individual thermal sensation model (future model) by modifying the latest past model based on the calculated estimated characteristic value SZest (step S605 in FIG. 16). Methods for correcting past models are preset by solution providers such as indoor environment evaluation/control and facility managers.

In this embodiment, the model correction parameter η is defined as the ratio between the estimated characteristic value SZest and the model characteristic value SZ1(1)=a(1) of the latest past model.
η=SZest/SZ1(1) (17)

Based on the model correction parameter η, the future model generator 22 corrects the function Mp(1) of the latest past model as follows.

Equation (18) uses the model correction parameter η It indicates to modify based on

In addition, in the correction of the past model of this embodiment, only the magnification correction for the PPVh axis is performed by the correction parameter η. Correction or magnification correction of the PMV axis may be performed.

Specifically, the future model generator 22 utilizes the characteristic quantity correction function Gpm_2 transmitted from the season-dependent management unit 21, and the model characteristic quantity SZ2 corresponding to the future outdoor thermal environment representative value Eotf = Gpm_2 (Eoft ) is calculated, and the calculated value is set as the estimated characteristic amount SZest_2. Further, the future model generation unit 22 uses the characteristic quantity correction function Gpm_3 to calculate the model characteristic quantity SZ3=Gpm_3 (Eoft) corresponding to the future outdoor thermal environment representative value Eotf, and converts the calculated value to the estimated characteristic quantity SZest_3 (step S604 in FIG. 16).

Then, for example, the difference between the estimated characteristic value SZest_2 and the model characteristic value SZ2(1)=e(1) of the latest past model is set as the model correction parameter δ.
δ=SZest_2-SZ2(1) (19)

Further, for example, the difference between the estimated characteristic quantities SZest_3 and SZest_2 (SZest_3-SZest_2) and the difference between the latest past model model characteristic quantities SZ3(1) and SZ2(1) (SZ3(1)-SZ2(1)) Let the ratio of the model correction parameter ξ be the model correction parameter ξ.
ξ = (SZest_3-SZest_2)/(SZ3(1)-SZ2(1))
(20)

Based on the model correction parameters η, δ, ξ, the future model generator 22 corrects the function Mp(1) of the latest past model as follows.

If the past model is corrected as in Equation (21), it is possible to correct the rising position of the reportee rate PPVh by the model correction parameter δ and to correct the magnification of the PMV axis by the model correction parameter ξ.

Thus, the processing of the future model generation unit 22 ends. According to this embodiment, by generating a future model by correcting the past model based on the outdoor thermal environment state in which the seasonal transition is reflected, it is possible to improve the estimation error due to the seasonal transition element. If the estimated PMV value of the estimated future date and time is input, the thermal sensation of the residents in the evaluation target space 101 at the estimated future date and time (the ratio of residents who report dissatisfaction with the thermal environment of the evaluation target space 101, such as "hot") It is possible to estimate the report rate PPVh). As a result, in the present embodiment, it is possible to make the estimation result of the thermal sensation of the resident closer to the actual situation.

[Second embodiment]
Next, a second embodiment of the invention will be described. Also in the present embodiment, the configuration of the thermal sensation model generating apparatus is the same as that of the first embodiment, so the description will be made using the reference numerals in FIG.
If learning data corresponding to a period of several days including short-term fluctuations is used, the reliability of the model cannot be maintained, so it is necessary to secure data for a certain period as learning data (integrated information). As described in the first embodiment, a learning data period width of about one month is appropriate, but in order to maintain the continuity of seasonal transitions, an overlap period is provided in which individual learning data periods overlap. In this embodiment, an example in which the overlap period is set to two weeks as a reliable fixed period will be described.

The data period management unit 14 of the thermal sensation model generating apparatus of the present embodiment includes data period management for determining the update timing of the individual thermal sensation model (past model) and the learning data period as in the first embodiment. The information is preconfigured by solution providers, such as indoor environment assessment/control, and facility managers.

The difference from the first embodiment is that the data period management information includes information on the overlap period width Lo in addition to information on the update timing of the individual thermal sensation model and information on the learning data period width.

FIGS. 17A and 17B are diagrams for explaining the learning data periods of the first and second embodiments. In addition, in FIGS. 17A and 17B, in order to make the explanation easier to understand, the date of the update timing is shown with 30 days for each month uniformly, and October 1, 2018 is the latest update timing. indicates the learning data period of each update timing going back to the update timing of August 1, 2018. In both FIGS. 17A and 17B, the update start date and time is 0:00 on August 1, 2018, and the learning data period is every 30 days. In the example of FIG. 17B, The overlap period width Lo in this embodiment is set to 15 days. In FIG. 17B of this embodiment, in which the overlap period width Lo is set for the learning data period with continuous update timings as shown in FIG. Correspondingly shortened.

In the example of this description corresponding to the first embodiment, as shown in FIG. 17A, the learning data period SW is 30 days starting at 00:00 on August 1 and ending at 23:59 on August 30 (here, each month is uniformly set to 30 days, so it corresponds to the end of August). be. Furthermore, the learning data period SW of the previous past model Mp(3) is set to start at 00:00 on July 1st and end on July 30th (in this case, each month is uniformly set to have 30 days, so the end of July). (equivalent to day) is 30 days ending at 23:59, and the learning data periods SW of past models Mp(2) and Mp(3) with consecutive update timings do not overlap.

On the other hand, in this example, as shown in FIG. The start timing is 00:00, and the end timing is 23:59 on August 30 (here, each month is uniformly set to 30 days, so it corresponds to the last day of August). is the same as the learning data period SW of the past model Mp(2) in FIG. 17B, but the learning data period SW of the previous past model Mp(4) in FIG. The timing includes the overlap period width Lo that overlaps with the learning data period SW of the past model Mp(3) in FIG. The learning data periods SW of the past models Mp(3) and Mp(4) with consecutive update timings overlap by the overlap period width Lo.

Other configurations are as described in the first embodiment. It should be noted that a period width of about one month is appropriate as the learning data period width in this embodiment, but in this case, in order to maintain the continuity of the seasonal transition, a reliable period width is overlapped between the learning data periods before and after. It is desirable to A long period of two weeks or longer is appropriate for the overlap period width Lo.

FIG. 18(A) shows the changes in the two-week average of outside air temperatures measured in a certain building from July to September 2017, and FIG. 18(B) shows the changes in the four-week average. The horizontal axes in FIGS. 18A and 18B are the number of elapsed weeks counted from the first week.

The overlap period is one week in the example of FIG. 18(A), and the overlap period is two weeks in the example of FIG. 18(B). As is clear from the comparison of FIGS. 18(A) and 18(B), short-term fluctuations are less likely to be included in the outside air temperature average value when the overlap period is two weeks and the average value for four weeks is taken. I understand.

The thermal sensation model generation device described in the first and second embodiments can be realized by a computer having a CPU (Central Processing Unit), a storage device, and an interface, and a program that controls these hardware resources. . A configuration example of this computer is shown in FIG.

The computer includes a CPU 200 , a storage device 201 and an interface device (hereinafter abbreviated as I/F) 202 . An environment measurement device or the like is connected to the I/F 202 . In such a computer, a program for realizing the thermal sensation model generation method of the present invention is stored in the storage device 201 . The CPU 200 executes the processes described in the first and second embodiments according to programs stored in the storage device 201 .

INDUSTRIAL APPLICABILITY The present invention can be applied to a technique for estimating a resident's thermal sensation at a future date and time.

1 past model generation unit 2 future model generation unit 10 data collection unit 11 living environment information storage unit 12 thermal sensation information storage unit 13 data integration unit 14 data period management unit 15 16 past model memory management unit 20 outdoor thermal environment information management unit 21 seasonal dependence management unit 22 future model generation unit.

Claims (10)

Data configured to store update timing information and learning data period information of a first model that models the relationship between a living environment index of an evaluation target space and a resident's thermal perception of the evaluation target space. period management department;
a first model generation unit configured to generate the first model based on the living environment index and the thermal sensation information of the resident in the learning data period when the update timing comes; ,
function of the first model based on the plurality of first models respectively generated by the plurality of learning data periods and the representative values of the past outdoor thermal environment corresponding to each of the plurality of learning data periods; a seasonal dependence manager configured to obtain a characteristic correction function indicating the relationship between the shape-determining model characteristic and a representative value of the outdoor thermal environment;
A future learning data period to be set for a future date and time for which the thermal sensation of the resident is to be estimated is determined, and a model characteristic value corresponding to a representative value of the outdoor thermal environment in this learning data period is calculated as the characteristic value. a second model generation unit configured to generate a second model obtained by calculating based on the correction function and modifying the latest first model based on the calculated model characteristic quantity. Characteristic thermal sensation model generation device. In the thermal sensation model generation device according to claim 1,
The data period management unit stores information on an overlap period in addition to information on the update timing and information on the learning data period,
The thermal sensation model generation device, wherein the plurality of learning data periods are set such that consecutive learning data periods overlap by a width of the overlapping period. The thermal sensation model generation device according to claim 1 or 2,
further comprising an outdoor thermal environment information management unit configured to store past outdoor thermal environment information of a building that includes the evaluation target space;
The season-dependent management unit acquires the past outdoor thermal environment information corresponding to each of the plurality of learning data periods from the outdoor thermal environment information management unit, and obtains the past outdoor thermal environment information corresponding to each of the plurality of learning data periods. Determining a representative value of the outdoor thermal environment to obtain the characteristic value correction function,
The second model generation unit acquires the past outdoor thermal environment information corresponding to the future learning data period from the outdoor thermal environment information management unit, and obtains the outdoor thermal environment information corresponding to the future learning data period. A thermal sensation model generation device characterized by determining a representative value of . In the thermal sensation model generation device according to any one of claims 1 to 3,
The second model generation unit corrects the latest first model based on the model characteristic quantity calculated based on the characteristic quantity correction function and the latest model characteristic quantity of the first model. A thermal sensation model generation device characterized by: In the thermal sensation model generation device according to any one of claims 1 to 4,
The thermal sensation model generating apparatus, wherein the living environment index is PMV, and the thermal sensation of the residents is a ratio of residents who report dissatisfaction with the thermal environment. Refer to the data period management unit that stores information on the update timing of the first model that models the relationship between the living environment index of the evaluation target space and the resident's thermal perception of the evaluation target space and information on the learning data period. a first step of generating the first model based on the living environment index and the thermal sensation information of the resident in the learning data period when the update timing comes;
function of the first model based on the plurality of first models respectively generated by the plurality of learning data periods and the representative values of the past outdoor thermal environment corresponding to each of the plurality of learning data periods; a second step of obtaining a characteristic correction function that indicates the relationship between the model characteristic that determines the shape and the representative value of the outdoor thermal environment;
A future learning data period to be set for a future date and time for which the thermal sensation of the resident is to be estimated is determined, and a model characteristic value corresponding to a representative value of the outdoor thermal environment in this learning data period is calculated as the characteristic value. and a third step of calculating based on the correction function and generating a second model obtained by correcting the latest first model based on the calculated model characteristic quantity. Method. In the thermal sensation model generation method according to claim 6,
The data period management unit stores information on an overlap period in addition to information on the update timing and information on the learning data period,
The thermal sensation model generating method, wherein the plurality of learning data periods are set such that consecutive learning data periods overlap by a width of the overlapping period. In the thermal sensation model generation method according to claim 6 or 7,
The second step refers to an outdoor thermal environment information management unit that stores past outdoor thermal environment information of a building that includes the evaluation target space, and stores the past outdoor thermal environment information corresponding to each of the plurality of learning data periods. acquiring thermal environment information from the outdoor thermal environment information management unit, determining representative values of past outdoor thermal environments corresponding to each of the plurality of learning data periods, and obtaining the characteristic amount correction function;
In the third step, the past outdoor thermal environment information corresponding to the future learning data period is acquired from the outdoor thermal environment information management unit, and a representative of the outdoor thermal environment corresponding to the future learning data period is obtained. A method of generating a thermal sensation model, comprising the step of determining a value. In the thermal sensation model generation method according to any one of claims 6 to 8,
The third step includes correcting the latest first model based on the model characteristic quantity calculated based on the characteristic quantity correction function and the latest model characteristic quantity of the first model. A thermal sensation model generation method characterized by: In the thermal sensation model generation method according to any one of claims 6 to 9,
A thermal sensation model generating method, wherein the living environment index is PMV, and the thermal sensation of the residents is a ratio of residents who report dissatisfaction with the thermal environment.

Images