JP7424558B2 - Control devices for personal air conditioning equipment, personal air conditioning systems and programs - Google Patents

Description

The present invention relates to a control device for personal air conditioning equipment, a personal air conditioning system, and a program.

Conventionally, a control device for air conditioning equipment uses a predictive model built using machine learning to control the air volume etc. of the air conditioning equipment based on conditions such as the temperature and humidity of the environment where the air conditioning equipment is installed and the air outlet temperature of the air conditioning equipment. (For example, see Patent Documents 1 and 2). In this conventional control device, when the user makes a declaration when the air volume etc. of the air conditioner do not match the user's preference, the data at the time of the declaration is stored. Then, machine learning is performed using this stored data, and the prediction model is updated.

Japanese Patent Application Publication No. 6-94292 Unexamined Japanese Patent Publication No. 2018-9770

However, in the conventional control device, for example, in order to update the prediction model to match the user's preferred airflow volume, a user's declaration is required, which is cumbersome. Further, in this type of control device, it is required to improve the accuracy of the blowout air volume in accordance with the user's preference.

The present invention has been made in view of the above circumstances, and it is possible to update the prediction model to match the user's preferred airflow volume without the user's notification, and to improve the accuracy of the airflow volume according to the user's preference. The purpose is to be able to improve.

A personal air conditioner control device according to a first aspect of the present invention is configured to control the temperature and humidity of the air outside the building when a user using the personal air conditioner installed in the building enters the building; Data acquisition for acquiring each data of the current elapsed time since the personal air conditioner was activated by the user, the current temperature and humidity of the air inside the building, and the current airflow volume of the personal air conditioner. an elapsed time determination unit that determines whether the elapsed time is within the specified time; a data storage unit that stores the data when the elapsed time is within the specified time; and the data storage unit. a machine learning unit that performs machine learning using the data stored in the user to update a predictive model for the user; a prediction processing unit that determines a setting value for the airflow volume of the personal air conditioner based on the temperature and humidity of the air inside the building and the temperature and humidity of the air inside the building when the personal air conditioner is activated; , a control unit that controls the amount of air blown from the personal air conditioner based on the elapsed time so that the amount of air blown from the personal air conditioner decreases or increases from the set value until the elapsed time reaches the specified time; and.

According to this personal air conditioner control device, the temperature and humidity of the air outside the building when the user enters the building, the current elapsed time since the personal air conditioner was started by the user, and the inside of the building. The data acquisition unit acquires data on the current temperature and humidity of the air and the current air volume of the personal air conditioner, and the elapsed time determination unit determines whether the elapsed time is within a specified time. . Further, when the elapsed time is within the specified time, data corresponding to this elapsed time is stored in the data storage section. Then, the machine learning unit performs machine learning using the data stored in the data storage unit, and updates the prediction model. Therefore, even without a user's notification, the prediction model can be updated to match the user's preferred airflow volume.

Furthermore, according to this personal air conditioner control device, as described above, machine learning is performed using data when the elapsed time is within the specified time, and the prediction model is updated. Therefore, when controlling the airflow volume using a prediction model when the elapsed time is within the specified time, the data when the elapsed time is within the specified time is reflected, so the accuracy of the airflow volume according to the user's preference is reflected. can be improved.

The personal air conditioner control device includes a first data storage unit as the data storage unit that stores the data when the elapsed time is within the specified time, and a first data storage unit as the data storage unit that stores the data when the elapsed time is within the specified time. a second data storage unit that stores the data as second data when the data exceeds 100, and the machine learning unit that executes machine learning using the first data and updates the first prediction model for the user. a second machine learning unit that performs machine learning using the second data and updates a second predictive model for the user; the personal air conditioner based on the temperature and humidity of the air outside the building when the user enters the building and the temperature and humidity of the air inside the building when the personal air conditioner is activated; The temperature of the air outside the building when the user enters the building is determined by using the first prediction processing unit as the prediction processing unit that determines the first set value of the blowout air volume, and the second prediction model. and a second prediction processing unit that determines a second set value of the airflow volume of the personal air conditioner based on the humidity and the temperature and humidity of the air inside the building when the personal air conditioner is activated. The controller is configured to change the airflow volume of the personal air conditioner from the first set value to the second set value based on the elapsed time until the elapsed time reaches the specified time. Then, after the elapsed time reaches the specified time, the personal air conditioner may control the amount of air blown from the personal air conditioner, and after the elapsed time reaches the specified time, the amount of air blown from the personal air conditioner may be maintained at the second set value.

According to this personal air conditioner control device, when the elapsed time exceeds the specified time, data corresponding to this elapsed time is stored as second data in the second data storage section. Then, the second machine learning section performs machine learning using the second data stored in the second data storage section, and updates the second prediction model. Therefore, even without a user's notification, the second prediction model can be updated to match the user's preferred airflow volume.

Moreover, according to this personal air conditioner control device, as described above, machine learning is performed using the second data when the elapsed time exceeds the specified time, and the second prediction model is updated. Therefore, when controlling the airflow volume using the second prediction model when the elapsed time exceeds the specified time, the data when the elapsed time exceeds the specified time is reflected, so the airflow volume corresponds to the user's preference. accuracy can be improved.

The control device for the personal air conditioner includes a first operation determination unit that determines that the user has performed an operation to change the blowout air volume of the personal air conditioner when the elapsed time is within the specified time; further comprising a second operation determination unit that determines that the user has performed an operation to change the blowout air volume of the personal air conditioner when the time exceeds the specified time; The first data storage unit stores the first data when the first operation determination unit determines that the change operation has been performed, and the second data storage unit stores the first data when the second operation determination unit determines that the change operation has been performed. The second data may be stored in the second data.

According to this personal air conditioner control device, when the elapsed time is within the specified time and the user performs an operation to change the blowout air volume of the personal air conditioner, the first data is stored in the first data storage unit. . Therefore, since the first data that reflects the user's preferences is stored in the first data storage unit, by machine learning the first prediction model using this first data, the elapsed time is within the specified time. In some cases, the blowout air volume can be controlled to better reflect the user's preferences. Similarly, when the elapsed time exceeds the specified time and the user performs an operation to change the airflow volume of the personal air conditioner, the second data is stored in the second data storage section. Therefore, the second data storage unit stores second data that reflects the user's preferences, and by using this second data to perform machine learning on the second prediction model, the elapsed time exceeds the specified time. In this case, the airflow volume can be controlled to more closely reflect the user's preferences.

In the personal air conditioner control device, the elapsed time determination section may change the prescribed time based on the temperature of air outside the building.

According to this control device for a personal air conditioner, the prescribed time is changed based on the temperature of the air outside the building, so it is possible to control the amount of air blown according to the temperature of the air outside the building.

In the above personal air conditioner control device, the data acquisition unit counts the elapsed time while the user information is held in the user information reading unit that reads user information about the user, and When the holding of the user information is canceled, the elapsed time may be reset.

According to this personal air conditioner control device, the elapsed time is counted while the user information about the user is held in the user information reading section, and when the user information reading section cancels the holding of the user information, Elapsed time is reset. Therefore, the elapsed time is counted until the user performs an operation to release the user information from the user information reading section, and the blowout air volume is controlled in accordance with this elapsed time. This can prevent the elapsed time from being reset against the user's will and the blowout air volume from being controlled against the user's will, for example, when the user temporarily leaves his or her seat.

A personal air conditioning system according to a second aspect of the present invention includes a personal air conditioning device installed in a building, an external sensor that detects the temperature and humidity of the air outside the building, and a personal air conditioning system that detects the temperature and humidity of the air inside the building. an internal sensor that detects humidity; and a control device that controls the personal air conditioner, and the control device detects air outside the building when a user using the personal air conditioner enters the building. temperature and humidity, the current elapsed time since the personal air conditioner was activated by the user, the current temperature and humidity of the air inside the building, and the current blowout air volume of the personal air conditioner. a data acquisition unit that acquires data; an elapsed time determination unit that determines whether the elapsed time is within the specified time; and a data storage unit that stores the data when the elapsed time is within the specified time. a machine learning unit that performs machine learning using the data stored in the data storage unit and updates a predictive model for the user; Based on the temperature and humidity of the air outside the building at the time and the temperature and humidity of the air inside the building when the personal air conditioner is started, set values for the airflow volume of the personal air conditioner are determined. a prediction processing unit that determines the amount of air blown from the personal air conditioner based on the elapsed time so that the amount of air blown from the personal air conditioner decreases or increases from the set value until the elapsed time reaches the specified time; A control unit that controls air volume.

According to this personal air conditioning system, the temperature and humidity of the air outside the building when the user enters the building, the current elapsed time since the personal air conditioning equipment was activated by the user, and the temperature and humidity of the air inside the building. The data acquisition unit acquires data on the current temperature and humidity and the current airflow volume of the personal air conditioner, and the elapsed time determination unit determines whether or not the elapsed time is within a specified time. Further, when the elapsed time is within the specified time, data corresponding to this elapsed time is stored in the data storage section. Then, the machine learning unit performs machine learning using the data stored in the data storage unit, and updates the prediction model. Therefore, even without a user's notification, the prediction model can be updated to match the user's preferred airflow volume.

Moreover, according to this personal air conditioning system, as described above, machine learning is performed using data when the elapsed time is within the specified time, and the prediction model is updated. Therefore, when controlling the airflow volume using a prediction model when the elapsed time is within the specified time, the data when the elapsed time is within the specified time is reflected, so the accuracy of the airflow volume according to the user's preference is reflected. can be improved.

The program according to the third aspect of the present invention is configured to calculate the temperature and humidity of the air outside the building when a user using the personal air conditioning equipment installed in the building enters the building, and the temperature and humidity of the air outside the building when the user using the personal air conditioning equipment installed in the building enters the building. a data acquisition step of acquiring each data of the current elapsed time since activation by the user, the current temperature and humidity of the air inside the building, and the current blowing air volume of the personal air conditioner, and the elapsed time. an elapsed time determining step of determining whether the elapsed time is within the specified time; a data storage step of storing the data in the data storage section when the elapsed time is within the specified time; and a data storage step of storing the data in the data storage section. a machine learning step of performing machine learning using the stored data to update a predictive model for the user; and using the predictive model to determine the exterior of the building when the user enters the building. a prediction processing step of determining a setting value for the airflow volume of the personal air conditioner based on the temperature and humidity of the air and the temperature and humidity of the air inside the building when the personal air conditioner is activated; a control step of controlling the air volume of the personal air conditioner based on the elapsed time so that the air volume of the personal air conditioner decreases or increases from the set value until the elapsed time reaches the specified time; This is a program that causes a computer to execute .

According to this program, the temperature and humidity of the air outside the building when the user entered the building, the current elapsed time since the personal air conditioning equipment was turned on by the user, and the current temperature and humidity of the air inside the building. Data on the temperature, humidity, and current airflow volume of the personal air conditioner are acquired in the data acquisition step, and it is determined in the elapsed time determination step whether or not the elapsed time is within a specified time. Further, when the elapsed time is within the specified time, data corresponding to this elapsed time is stored in the data storage section. Then, in the machine learning step, machine learning is performed using the data stored in the data storage unit, and the prediction model is updated. Therefore, even without a user's notification, the prediction model can be updated to match the user's preferred airflow volume.

Moreover, according to this program, as described above, machine learning is performed using data when the elapsed time is within the specified time, and the prediction model is updated. Therefore, when controlling the airflow volume using a prediction model when the elapsed time is within the specified time, the data when the elapsed time is within the specified time is reflected, so the accuracy of the airflow volume according to the user's preference is reflected. can be improved.

As detailed above, according to the present invention, the prediction model can be updated to match the user's preferred airflow volume without the user's notification, and the accuracy of the airflow volume relative to the user's preference can be improved. be able to.

1 is an overall configuration diagram of a personal air conditioning system according to an embodiment of the present invention, and is a diagram showing a control device in terms of hardware configuration. FIG. 1 is an overall configuration diagram of a personal air conditioning system according to an embodiment of the present invention, and is a diagram showing a control device using functional blocks for data recording processing. FIG. 1 is an overall configuration diagram of a personal air conditioning system according to an embodiment of the present invention, and is a diagram showing a control device using functional blocks for control processing. FIG. 3 is a flowchart showing the flow of data recording processing executed by the control device. 3 is a flowchart showing the flow of control processing executed by the control device. FIG. 3 is a diagram illustrating an example of a control pattern for controlling the amount of air blown from a personal air conditioner during cooling.

An embodiment of the present invention will be described below.

FIG. 1 shows the overall configuration of a personal air conditioning system 10 according to an embodiment of the present invention. As shown in FIG. 1, a personal air conditioning system 10 according to the present embodiment is installed in a building 12 such as a building containing an office, and includes a personal air conditioning device 14, an external sensor 16, and an internal sensor 18. , a user information reader 20 , and a control device 22 .

The personal air conditioner 14, the external sensor 16, the internal sensor 18, and the user information reader 20 are communicably connected to the control device 22. Furthermore, the building 12 is equipped with an admission management system 24. The admission management system 24 is communicably connected to the control device 22. For example, the entrance management system 24 reads the user information recorded on the entrance card owned by the user 26 when the user 26 enters and leaves the museum, outputs the entrance information when the user 26 enters the museum, and outputs the entrance information when the user 26 leaves the museum. Output exit information.

Personal air conditioner 14 is installed inside building 12. This personal air conditioner 14 includes a seat 28 for use by a user 26, and is used individually by the user 26. This personal air conditioner 14 is configured to blow out conditioned air 30 to a user 26 seated on a seat 28.

This personal air conditioner 14 has an on/off function that can be switched to start and stop, a mode switching function that can switch between automatic mode (wellness mode) and manual mode, and an air conditioner that can set the blowout temperature and blowout air volume of the conditioned air 30. It has a setting function. The personal air conditioner 14 has an operation panel 32, and the operation panel 32 is provided with operation sections such as buttons corresponding to each of the above functions.

When the operation section for the on/off function is operated on the operation panel 32, the operation panel 32 outputs on/off information corresponding to this operation, and when the operation section for the mode switching function is operated on the operation panel 32, this operation is output. The configuration outputs mode switching information corresponding to the air conditioner, and when the operation section of the air conditioning setting function is operated on the operation panel 32 and the outlet temperature and outlet air volume are changed, outputs data on the changed outlet temperature and outlet air volume. It is said that

Note that a user terminal 34 such as a smartphone, a tablet terminal, or a notebook computer owned by the user 26 may be able to communicate with the personal air conditioner 14, and this user terminal 34 may be used instead of the operation panel 32. Further, instead of the on/off function of the operation panel 32, a seating sensor that detects that the user 26 is seated may be used, and the seating sensor may output on/off information depending on whether the user 26 is seated.

The external sensor 16 is provided outside the building 12 (for example, near the entrance of the building 12). This external sensor 16 includes an external temperature sensor 36 and an external humidity sensor 38. The external temperature sensor 36 detects the temperature of the air outside the building 12 (outside air temperature), and the external humidity sensor 38 detects the humidity of the air outside the building 12 (outside air humidity). This external sensor 16 is configured to output data on the temperature and humidity of the air outside the building 12 detected by an external temperature sensor 36 and an external humidity sensor 38.

The internal sensor 18 is provided inside the building 12 (for example, near the personal air conditioner 14). This internal sensor 18 has an internal temperature sensor 40 and an internal humidity sensor 42. The internal temperature sensor 40 detects the temperature of the air inside the building 12 (inside air temperature), and the internal humidity sensor 42 detects the humidity of the air inside the building 12 (inside air humidity). This internal sensor 18 is configured to output data on the temperature and humidity of the air inside the building 12 detected by the internal temperature sensor 40 and internal humidity sensor 42.

The user information reader 20 can communicate with the user terminal 34. The user information reader 20 reads user information stored in the user terminal 34 when the user 26 performs a read request operation such as bringing the user terminal 34 close to the user information reader 20. , and is configured to output the read user information. The user terminal 34 that communicates with the user information reader 20 may be, for example, an ID card or a wearable terminal in addition to a smartphone, a tablet terminal, or a notebook computer. Further, instead of the on/off function of the operation panel 32 described above, when user information is read by the user information reader 20, the on/off information may be output from the user information reader 20.

The control device 22 is an example of a "computer". This control device 22 controls the airflow volume of the personal air conditioning device 14 based on information and data output from the personal air conditioning device 14, external sensor 16, internal sensor 18, user information reader 20, and entrance management system 24. It is. FIG. 1 shows the hardware configuration of the control device 22. As shown in FIG.

This control device 22 has an input/output interface 44, a CPU (Central Processing Unit) 46, a ROM (Read Only Memory) 48, a RAM (Random Access Memory) 50, and a nonvolatile memory 52 as a hardware configuration. are doing.

The input/output interface 44, the CPU 46, the ROM 48, the RAM 50, and the nonvolatile memory 52 are connected via a bus 54 so as to be able to communicate with each other. A program 56 for controlling the amount of air blown from the personal air conditioner 14 is stored in the nonvolatile memory 52 . This program 56 has a process for data recording processing and a process for control processing.

FIG. 2 shows functional blocks of the control device 22 that executes data recording processing. As shown in FIG. 2, the control device 22 includes a data acquisition section 60, a data recording section 62, a mode determination section 64, an elapsed time determination section 66, and a first Operation determination unit 68, second operation determination unit 70, first data storage unit 72, second data storage unit 74, first machine learning unit 76, second machine learning unit 78, and first prediction model It has a storage unit 80 and a second prediction model storage unit 82.

The data acquisition unit 60, the mode determination unit 64, the elapsed time determination unit 66, the first operation determination unit 68, the second operation determination unit 70, the first machine learning unit 76, and the second machine learning unit 78 are , is realized by the CPU 46 shown in FIG. 1 executing the program 56. On the other hand, the data recording section 62, the first data storage section 72, the second data storage section 74, the first predictive model storage section 80, and the second predictive model storage section 82 are nonvolatile memories shown in FIG. 52. The first data storage unit 72 is an example of a “data storage unit”, and the first machine learning unit 76 is an example of a “machine learning unit”.

FIG. 3 shows functional blocks of the control device 22 that executes control processing. As shown in FIG. 3, the control device 22 includes the above-described data acquisition section 60, data recording section 62, elapsed time determination section 66, first prediction model storage section 80, and second prediction model storage section 80 as functional sections that execute control processing. In addition to the prediction model storage section 82, it includes a user determination section 84, a first prediction processing section 86, a second prediction processing section 88, a control section 90, and a default processing section 92.

The user determination section 84, the first prediction processing section 86, the second prediction processing section 88, the control section 90, and the default processing section 92 are realized by the CPU 46 shown in FIG. 1 executing the program 56. . The first prediction processing unit 86 is an example of a “prediction processing unit”.

Next, a method of controlling the personal air conditioner 14 according to this embodiment will be explained.

In the following description, reference will be made to the above-mentioned FIGS. 1 to 3 as appropriate for the configuration of the personal air conditioning system 10. FIG. 4 shows the flow of data recording processing executed by the control device 22. The data recording process is executed when the personal air conditioner 14 is activated. In this data recording process, the following steps S1 to S12 are repeatedly executed.

In step S1, the current elapsed time since the personal air conditioner 14 was started by the user 26, the current temperature and humidity of the air inside the building 12, and the current air volume and temperature of the personal air conditioner 14 are determined. Each piece of data is acquired by the data acquisition unit 60. Data on the current elapsed time since the personal air conditioner 14 was activated by the user 26 is acquired using the clock function of the control device 22.

Note that the control device 22 determines whether the user 26 has entered the building 12 based on the entrance information output from the entrance management system 24 before the personal air conditioner 14 is activated (before step S1 described above is executed). Detecting this, data on the temperature and humidity of the air outside the building 12 when the user 26 enters the building 12 is acquired in advance, and this data is temporarily stored.

In step S1, the temperature and humidity of the air outside the building 12 when the user 26 enters the building 12, the current elapsed time since the personal air conditioner 14 was started by the user 26, and the temperature and humidity of the air outside the building 12 are determined. Data on the current temperature and humidity of the internal air, and the current air volume and temperature of the personal air conditioner 14 are acquired as a set by the data acquisition unit 60. Further, in this step S1, the data acquisition unit 60 acquires user information and mode switching information in addition to the acquired data described above. This step S1 is an example of a "data acquisition step."

In step S2, the data acquired by the data acquisition section 60 is recorded in the data recording section 62 in association with user information and mode switching information.

In step S3, the mode of the personal air conditioner 14 is determined by the mode determining section 64. Here, when the mode of the personal air conditioner 14 is the automatic mode (wellness mode), the process returns to step S1 described above, and when the mode of the personal air conditioner 14 is the manual mode, the process moves to step S4.

In step S4, the elapsed time determining unit 66 determines whether the current elapsed time since the personal air conditioner 14 was started by the user 26 is within a specified time. This step S4 is an example of an "elapsed time determination step."

As a result of extensive studies, the present inventors have found that even if the user 26 is seated in a high metabolic state, the metabolic state of the user 26 returns to a normal stable state after 30 minutes have passed since sitting. Ta. The prescribed time corresponds to the time required for the metabolic state of the user 26 to return to a normal stable state. Based on the above knowledge, in this embodiment, the specified time is set to 30 minutes, as an example. Here, if the elapsed time is within the specified time, the process moves to step S5.

In step S5, the first operation determination unit 68 determines whether or not the air volume of the personal air conditioner 14 has been changed during the current elapsed time. Whether or not the airflow volume of the personal air conditioner 14 has been changed is determined by comparing the airflow volume at the current elapsed time with the past airflow volume recorded in the data recording unit 62. This step S5 is an example of a "first operation determination step."

Here, if the air volume of the personal air conditioner 14 has not been changed, the process returns to step S1, and if the air volume of the personal air conditioner 14 has been changed, the process moves to step S6.

In step S6, the data acquired in step S1 described above, that is, the data when the current elapsed time is within the specified time, is stored as first data in the first data storage unit 72. This step S6 is an example of a "data storage step" and a "first data storage step."

In step S7, the first machine learning unit 76 performs machine learning using the first data stored in the first data storage unit 72, and updates the first prediction model for the user 26. This step S7 is an example of a "machine learning step" and a "first machine learning step."

For example, support vector regression (SVR), a neural network, or the like is used as the first prediction model. The first prediction model is based on the temperature and humidity of the air outside the building 12 when the user 26 enters the building 12, the current elapsed time since the personal air conditioner 14 was activated by the user 26, and the temperature and humidity of the air outside the building 12 when the user 26 enters the building 12. The current temperature and humidity of the internal air and the current blowout temperature of the personal air conditioner 14 are used as explanatory variables, the current blowout air volume of the personal air conditioner 14 is used as the target variable, and the explanatory variables and the target variable are used as training data for learning. be done.

In step S8, the first predictive model updated by the first machine learning unit 76 is stored in the first predictive model storage unit 80.

On the other hand, if the elapsed time exceeds the specified time in step S4, the process moves to step S9.

In step S9, the second operation determination unit 70 determines whether or not the air volume of the personal air conditioner 14 has been changed during the current elapsed time. Whether or not the airflow volume of the personal air conditioner 14 has been changed is determined by comparing the airflow volume at the current elapsed time with the past airflow volume recorded in the data recording unit 62. This step S9 is an example of a "second operation determination step."

Here, if the air volume of the personal air conditioner 14 has not been changed, the process returns to step S1, and if the air volume of the personal air conditioner 14 has been changed, the process moves to step S10.

Note that in step S9 for the first time after the current elapsed time exceeds the specified time, the process moves to step S10 regardless of whether or not there is a change operation.

In step S10, the data acquired in step S1 described above, that is, the data when the current elapsed time exceeds the specified time, is stored in the second data storage section 74 as second data. This step S10 is an example of a "second data storage step."

In step S11, the second machine learning unit 78 performs machine learning using the second data stored in the second data storage unit 74, and updates the second prediction model for the user 26. This step S11 is an example of a "second machine learning step."

For example, support vector regression (SVR), a neural network, or the like is used as the second prediction model. Similar to the first prediction model, the second prediction model calculates the temperature and humidity of the air outside the building 12 when the user 26 enters the building 12, and the current temperature and humidity after the personal air conditioner 14 is activated by the user 26. The explanatory variables are the elapsed time of and the objective variable as training data.

In step S12, the second predictive model updated by the second machine learning unit 78 is stored in the second predictive model storage unit 82.

The above steps S1 to S12 are executed every time data is acquired by the data acquisition unit 60.

FIG. 5 shows the flow of control processing executed by the control device 22. The control process is executed when automatic mode (wellness mode) is selected. In this control process, the following steps S21 to S26 are repeatedly executed.

In step S21, data on the current elapsed time since the personal air conditioner 14 was activated by the user 26 is acquired by the data acquisition unit 60. Data on the current elapsed time since the personal air conditioner 14 was activated by the user 26 is acquired using the clock function of the control device 22.

Note that the control device 22 determines whether the user 26 has entered the building 12 based on the entrance information output from the entrance management system 24 before the personal air conditioner 14 is activated (before step S21 described above is executed). Detecting this, data on the temperature and humidity of the air outside the building 12 when the user 26 enters the building 12 is acquired in advance, and this data is temporarily stored. Furthermore, when the personal air conditioner 14 is started, the control device 22 receives data on the temperature and humidity of the air inside the building 12 outputted from the internal sensor 18 and the air outlet temperature outputted from the personal air conditioner 14. and temporarily stores it.

Then, in step S21, the temperature and humidity of the air outside the building 12 when the user 26 enters the building 12, the current elapsed time since the personal air conditioner 14 was started by the user 26, and the personal air conditioner 14 is activated by the user 26, and the temperature and humidity of the air inside the building 12 when the personal air conditioner 14 is activated by the user 26 are set. The data is acquired by the data acquisition unit 60.

Among the data acquired in step S21, the temperature and humidity of the air outside the building 12 when the user 26 enters the building 12, and the inside of the building 12 when the personal air conditioner 14 is activated by the user 26. The data on the temperature and humidity of the air, and the blowout temperature of the personal air conditioner 14 when the personal air conditioner 14 is started by the user 26 are treated as data when the personal air conditioner 14 is started. Further, in this step S21, user information is acquired by the data acquisition unit 60 in addition to the above-mentioned acquired data.

In step S22, the user determination unit 84 reads the user information acquired by the data acquisition unit 60, and based on this user information, from the data recorded in the data recording unit 62, it is determined that the user 26 has It is determined whether the person has ever performed an operation to change the amount of air blown from the device 14.

Here, in step S22, if it is determined that the user 26 is not a person who has changed the airflow volume of the personal air conditioner 14 in the past, the process moves to step S23.

In step S23, default processing is executed by the default processing unit 92. That is, as a default process, the amount of air blown from the personal air conditioner 14 is controlled to a predetermined amount of air (default value).

On the other hand, if it is determined in step S22 that the user 26 has changed the airflow volume of the personal air conditioner 14 in the past, the process moves to step S24.

In step S24, the elapsed time determining unit 66 determines whether or not the personal air conditioner 14 is activated by the user 26. Here, if it is the time of startup, the process moves to steps S25-1 and S25-2.

In step S25-1, the first prediction processing unit 86 reads out the first prediction model stored in the first prediction model storage unit 80, and using this first prediction model, the Among the data, the temperature and humidity of the air outside the building 12 when the user 26 enters the building 12, and the temperature and humidity of the air inside the building 12 when the personal air conditioning device 14 is activated by the user 26. , the first setting value SP1 (Fig. 6) is determined. The first set value SP1 will be described later. This step S25-1 is an example of a "prediction processing step" and a "first prediction processing step."

In step S25-2, the second prediction processing unit 88 reads out the second prediction model stored in the second prediction model storage unit 82, and using this second prediction model, the Among the data, the temperature and humidity of the air outside the building 12 when the user 26 enters the building 12, and the temperature and humidity of the air inside the building 12 when the personal air conditioning device 14 is activated by the user 26. , a second setting value SP2 of the airflow volume of the personal air conditioning device 14 (Fig. 6) is determined. The second set value SP2 will be described later. This step S25-2 is an example of a "second prediction processing step."

Here, FIG. 6 shows an example of a control pattern for controlling the amount of air blown from the personal air conditioner 14 during cooling. The first set value SP1 determined in step S25-1 described above is, for example, the upper limit of the airflow volume, and the second set value SP2 determined in step S25-2 described above is, as an example, the upper limit of the airflow volume. is the lower limit of

In this control pattern, for example, before the elapsed time reaches the specified time T, the air volume blown from the personal air conditioner 14 is gradually reduced from the first set value SP1 to the second set value SP2. The time until the elapsed time reaches the specified time T is divided into a first time t1 and a plurality of second times t2.

The first time t1 is the first time interval after the personal air conditioner 14 is activated, and the plurality of second times t2 are time intervals after the first time t1. For example, the specified time T is 30 minutes, the first time t1 is 10 minutes, and the second time t2 is 3 minutes. The amount of decrease in the blowout air volume is determined according to the difference between the first set value SP1 and the second set value SP2, the number of the plurality of second times t2, and the specified time T.

In step S26, the control unit 90 controls the amount of air blown from the personal air conditioner 14 based on the control pattern shown in FIG.

That is, in the first step S26, since the elapsed time acquired in step S21 is the time when the personal air conditioner 14 is activated, the control value of the airflow volume of the personal air conditioner 14 is determined to be the first setting value SP1.

Further, after starting the personal air conditioner 14, step S26 is executed after step S24, and in the second and subsequent steps S26, based on the elapsed time acquired in step S21 and the control pattern shown in FIG. , a control value for the air volume of the personal air conditioner 14 is determined.

Then, until the elapsed time reaches the specified time T, the elapsed time obtained in step S21 is set so that the air volume of the personal air conditioner 14 is gradually reduced from the first set value SP1 to the second set value SP2. Based on this, the amount of air blown from the personal air conditioner 14 is controlled. On the other hand, after reaching the specified time T, control is performed to maintain the airflow volume of the personal air conditioner 14 at the second set value SP2. This step S26 is an example of a "control step."

Next, the functions and effects of this embodiment will be explained.

As detailed above, according to the present embodiment, the temperature and humidity of the air outside the building 12 when the user 26 enters the building 12, and the current temperature and humidity after the personal air conditioner 14 is activated by the user 26. The data acquisition unit 60 acquires each data of the elapsed time, the current temperature and humidity of the air inside the building 12, and the current blowout air volume and blowout temperature of the personal air conditioner 14, and determines that the elapsed time is within a specified time. The elapsed time determining section 66 determines whether or not there is.

Furthermore, when the elapsed time is within the specified time, data corresponding to this elapsed time is stored in the first data storage section 72 as first data. Then, the first machine learning unit 76 executes machine learning using the first data stored in the first data storage unit 72, and updates the first prediction model. Therefore, even if the user 26 does not make a declaration, the first prediction model can be updated to match the user's 26 preferred airflow volume.

Moreover, as described above, machine learning is performed using the first data when the elapsed time is within the specified time, and the first prediction model is updated. Therefore, when controlling the blowout air volume using the first prediction model when the elapsed time is within the specified time, the first data when the elapsed time is within the specified time is reflected, so the preference of the user 26 is It is possible to improve the accuracy of the blowout air volume.

Similarly, when the elapsed time exceeds the specified time, data corresponding to this elapsed time is stored in the second data storage section 74 as second data. Then, the second machine learning unit 78 performs machine learning using the second data stored in the second data storage unit 74, and updates the second prediction model. Therefore, even if there is no notification from the user 26, the second prediction model can be updated to match the airflow rate desired by the user 26.

Moreover, as described above, machine learning is performed using the second data when the elapsed time exceeds the specified time, and the second prediction model is updated. Therefore, when controlling the blowout air volume using the second prediction model when the elapsed time exceeds the specified time, the data when the elapsed time exceeds the specified time is reflected, so that the blowout according to the user 26's preference is The accuracy of air volume can be improved.

Further, according to the present embodiment, when the user 26 performs an operation to change the airflow volume of the personal air conditioner 14 when the elapsed time is within the specified time, the first data is stored in the first data storage unit 72. Ru. Therefore, the first data storage unit 72 stores first data that reflects the preferences of the user 26, and by using this first data to perform machine learning on the first prediction model, the elapsed time is determined to be within the specified time. If it is within the range, the air volume can be controlled to more closely reflect the preferences of the user 26.

Similarly, when the elapsed time exceeds the specified time and the user 26 performs an operation to change the airflow volume of the personal air conditioner 14, the second data is stored in the second data storage section 74. Therefore, the second data storage unit 74 stores second data that reflects the preferences of the user 26, so that by machine learning the second prediction model using this second data, the elapsed time can be set to a predetermined time. If the amount of air exceeds the amount of air, the air volume can be controlled to more closely reflect the user's 26 preferences.

Next, a modification of the first embodiment will be described.

In the above embodiment, the blowout temperature of the personal air conditioner 14 is changeable, but the blowout temperature does not need to be changeable.

In the above embodiment, the personal air conditioner 14 outputs the air outlet temperature data, and this air outlet temperature data is used in the data recording process and control process described above. It does not have to be used for control processing. Furthermore, the personal air conditioner 14 does not need to output the data on the blowout temperature.

In the above embodiment, only the volume of air blown from the personal air conditioner 14 is controlled, but the temperature of the air blown may be controlled in addition to the volume of air blown.

In the above embodiment, the airflow volume of the personal air conditioner 14 is decreased stepwise from the first set value SP1 to the second set value SP2, but is continuously decreased from the first set value SP1 to the second set value SP2. may be done.

In the embodiment described above, the above-described data recording process and control process are applied when controlling the blowout air volume during cooling of the personal air conditioner 14, but the above-described data recording process and control process are applied when controlling the blowout air volume during heating of the personal air conditioner 14. Data recording processing and control processing may be applied. Further, during heating, the blowout air volume may be decreased or increased from the first set value SP1 to the second set value SP2.

In the embodiment described above, the specified time that is compared with the elapsed time by the elapsed time determining section 66 is set to 30 minutes, but the specified time may be other than 30 minutes.

In the above embodiment, the prescribed time is fixed, but the elapsed time determination unit 66 may change the prescribed time based on the temperature of the air outside the building 12, for example. That is, the specified time may be increased or decreased depending on the temperature of the air outside the building 12, or may be increased or decreased depending on the absolute value of the difference between the temperature of the air outside the building 12 and the temperature of the air inside the building 12. It may be increased or decreased.

With this configuration, the prescribed time is changed based on the temperature of the air outside the building 12, so the amount of air blown can be controlled in accordance with the temperature of the air outside the building 12.

In the embodiment described above, the data acquisition unit 60 counts the elapsed time while the user 26 enters the building 12 and the user information is held in the entrance management system 24, and counts the elapsed time while the user 26 exits the building 12 and manages the entrance management system. If the system 24 releases user information, the elapsed time may be reset.

Similarly, when the user information is read by the user information reader 20 due to a reading request operation by the user 26, the data acquisition unit 60 determines the elapsed time while the user information is held by the user information reader 20. The elapsed time may be reset when the user information reader 20 releases the user information from the user information reader 20 due to a reading release operation by the user 26.

With this configuration, the elapsed time is counted until the user 26 performs an operation to release the user information using the admission management system 24 or the user information reader 20, and the blowout air volume is adjusted according to this elapsed time. control is performed. This prevents the elapsed time from being reset against the user's will and the airflow volume being controlled against the user's will, for example, when the user 26 temporarily leaves his or her seat. can. In this example, the admission management system 24 and the user information reader 20 each correspond to an example of a "user information reader".

Note that among the plurality of modifications described above, modifications that can be combined may be combined.

Although one embodiment of the present invention has been described above, the present invention is not limited to the above, and it is of course possible to implement various modifications other than the above without departing from the spirit thereof. It is.

10 Personal air conditioning system 12 Building 14 Personal air conditioning equipment 16 External sensor 18 Internal sensor 20 User information reader (an example of a user information reader)
22 Control device (example of computer)
24 Admission management system (example of user information reading section)
26 User 56 Program 60 Data acquisition unit 62 Data recording unit 64 Mode determination unit 66 Elapsed time determination unit 68 First operation determination unit 70 Second operation determination unit 72 First data storage unit (an example of a data storage unit)
74 Second data storage unit 76 First machine learning unit (an example of a machine learning unit)
78 Second machine learning section 80 First prediction model storage section 82 Second prediction model storage section 84 User determination section 86 First prediction processing section (an example of a prediction processing section)
88 Second prediction processing section 90 Control section

Claims (7)

The temperature and humidity of the air outside the building when a user using a personal air conditioner installed in the building enters the building, and the current elapsed time since the personal air conditioner was activated by the user. and a data acquisition unit that acquires data on the current temperature and humidity of the air inside the building, and the current airflow volume of the personal air conditioner;
an elapsed time determination unit that determines whether the elapsed time is within a specified time;
a data storage unit that stores the data when the elapsed time is within the specified time;
a machine learning unit that performs machine learning using the data stored in the data storage unit and updates a prediction model for the user;
Using the predictive model, determine the temperature and humidity of the air outside the building when the user enters the building, and the temperature and humidity of the air inside the building when the personal air conditioning device is activated. a prediction processing unit that determines a set value of the blowout air volume of the personal air conditioner based on the information;
a control unit that controls the air volume of the personal air conditioner based on the elapsed time so that the air volume of the personal air conditioner decreases or increases from the set value until the elapsed time reaches the specified time; ,
A control device for personal air conditioning equipment. a first data storage unit as the data storage unit that stores the data when the elapsed time is within the specified time as first data;
a second data storage unit that stores the data when the elapsed time exceeds the specified time as second data;
a first machine learning unit as the machine learning unit that executes machine learning using the first data and updates a first prediction model for the user;
a second machine learning unit that performs machine learning using the second data and updates a second prediction model for the user;
Using the first prediction model, the temperature and humidity of the air outside the building when the user enters the building, and the temperature and humidity of the air inside the building when the personal air conditioning device is activated. a first prediction processing unit as the prediction processing unit that determines a first set value of the blowing air volume of the personal air conditioner based on the above;
Using the second predictive model, the temperature and humidity of the air outside the building when the user enters the building and the temperature and humidity of the air inside the building when the personal air conditioning device is activated. a second prediction processing unit that determines a second set value of the blowout air volume of the personal air conditioner based on;
Equipped with
The control unit includes:
Based on the elapsed time, the airflow of the personal air conditioner is changed from the first set value to the second set value until the elapsed time reaches the specified time. Control the air volume,
After the elapsed time reaches the specified time, control is performed to maintain the airflow volume of the personal air conditioner at the second set value;
The control device for personal air conditioning equipment according to claim 1. a first operation determination unit that determines that the user has performed an operation to change the airflow volume of the personal air conditioner when the elapsed time is within the specified time;
a second operation determination unit that determines that the user has performed an operation to change the airflow volume of the personal air conditioner when the elapsed time exceeds the specified time;
Furthermore,
The first data storage unit stores the first data when the first operation determination unit determines that the change operation has been performed,
The second data storage unit stores the second data when the second operation determination unit determines that the change operation has been performed.
The control device for personal air conditioning equipment according to claim 2. The elapsed time determination unit changes the prescribed time based on the temperature of the air outside the building.
A control device for a personal air conditioner according to any one of claims 1 to 3. The data acquisition unit counts the elapsed time while the user information is held by a user information reading unit that reads user information about the user, and when the user information reading unit releases the holding of the user information. in which case, resetting said elapsed time;
A control device for a personal air conditioner according to any one of claims 1 to 4. Personal air conditioning equipment installed in the building,
an external sensor that detects the temperature and humidity of the air outside the building;
an internal sensor that detects the temperature and humidity of the air inside the building;
a control device that controls the personal air conditioner;
Equipped with
The control device includes:
The temperature and humidity of the air outside the building when the user using the personal air conditioning equipment enters the building, the current elapsed time since the personal air conditioning equipment was activated by the user, and the temperature and humidity of the air outside the building when the user using the personal air conditioning equipment enters the building; a data acquisition unit that acquires data on the current temperature and humidity of internal air and the current air volume of the personal air conditioner;
an elapsed time determination unit that determines whether the elapsed time is within a specified time;
a data storage unit that stores the data when the elapsed time is within the specified time;
a machine learning unit that performs machine learning using the data stored in the data storage unit and updates a prediction model for the user;
Using the predictive model, determine the temperature and humidity of the air outside the building when the user enters the building, and the temperature and humidity of the air inside the building when the personal air conditioning device is activated. a prediction processing unit that determines a set value of the blowout air volume of the personal air conditioner based on the information;
a control unit that controls the air volume of the personal air conditioner based on the elapsed time so that the air volume of the personal air conditioner decreases or increases from the set value until the elapsed time reaches the specified time; ,
Equipped with
Personal air conditioning system. The temperature and humidity of the air outside the building when a user using a personal air conditioner installed in the building enters the building, and the current elapsed time since the personal air conditioner was activated by the user. and a data acquisition step of acquiring data on the current temperature and humidity of the air inside the building, and the current airflow volume of the personal air conditioner,
an elapsed time determining step of determining whether the elapsed time is within a specified time;
a data storage step of storing the data in a data storage unit when the elapsed time is within the specified time;
a machine learning step of performing machine learning using the data stored in the data storage unit and updating a predictive model for the user;
Using the predictive model, determine the temperature and humidity of the air outside the building when the user enters the building, and the temperature and humidity of the air inside the building when the personal air conditioning device is activated. a prediction processing step of determining a set value of the blowout air volume of the personal air conditioner based on the above;
a control step of controlling the air volume of the personal air conditioner based on the elapsed time so that the air volume of the personal air conditioner decreases or increases from the set value until the elapsed time reaches the specified time; ,
A program that causes a computer to execute