Method and system for variable air conditioning control through prediction of occupancy probability based on occupancy history

Abstract

Provided are a method and system for variable air conditioning control through prediction of an occupancy probability based on an occupancy history. An air conditioning control method, according to an embodiment of the present invention, comprises: calculating an occupancy prediction probability which is an occupancy probability after a predetermined time; increasing a standby count when the occupancy prediction probability exceeds a first threshold value in a state in which there is no occupant currently present; and initializing the standby count and turning off an air conditioner when the standby count exceeds a current maximum standby count. Accordingly, both comfort of an occupant and energy saving may be adaptively satisfied while controlling pre-air conditioning through the occupancy probability predicted on the basis of the occupancy history.

Description

Variable air conditioning control method and system using occupancy probability prediction based on occupancy history

[0001] The present invention relates to air conditioning control technology, and more particularly, to a method and system for variably controlling air conditioning by predicting the probability of occupancy based on past occupancy history.

[0002] Conventional air conditioning control technology is mostly a simple system operation that turns the air conditioning on and off at different times during the week based on a schedule, and when it is in operation, it is mostly controlled to produce a constant output without any fluctuations in the middle.

[0003] This technology cannot respond to changing circumstances, so the system must continue to operate without recognizing changes in the heating and cooling system, and it is impossible to control the air conditioning according to the conditions of each room.

[0004] Moreover, it is not only impossible to determine whether an individual room is occupied, but also the prediction relies on people's experience to set the schedule.

[0005] The present invention has been devised to solve the above problems, and the purpose of the present invention is to provide a variable air conditioning control method and system that reasonably satisfies both the comfort of occupants and energy savings while controlling air conditioning through an occupancy probability predicted based on occupancy history.

[0006] In order to achieve the above object, an air conditioning control method according to one embodiment of the present invention comprises the steps of: determining whether there is currently an occupant; calculating an occupancy prediction probability, which is a probability of occupancy after a predetermined time; increasing a standby count when the occupancy prediction probability exceeds a first threshold in a state where there is currently no occupant; and initializing the standby count and turning off the air conditioner when the standby count exceeds a current maximum standby count.

[0007] The current maximum waiting count may vary depending on the predicted probability of occupancy. The current maximum waiting count is:

[0008] Current maximum waiting count = maximum waiting count × occupancy prediction probability

[0009] It can be calculated using the above formula.

[0010] And the air conditioning control method according to an embodiment of the present invention may further include a step of controlling the air conditioner based on the current temperature if the standby count is less than or equal to the maximum standby count.

[0011] The air conditioning control method according to an embodiment of the present invention may include an air conditioner control step, wherein if the standby count is less than or equal to the maximum standby count, the air conditioner control step may include: a step of determining whether the current temperature matches the set temperature; a step of initializing the standby count and turning off the air conditioner if the current temperature matches the set temperature; and a step of turning on and controlling the air conditioner if the current temperature does not match the set temperature.

[0012] The air conditioning control method according to an embodiment of the present invention may further include a step of initializing a standby count and turning off the air conditioner when the occupancy prediction probability is lower than a first threshold value in a state where there is no occupant currently present.

[0013] The air conditioning control method according to an embodiment of the present invention may further include a step of initializing a standby count and turning off the air conditioner if the occupancy prediction probability is lower than a second threshold value that is lower than a first threshold value in a state where there is currently an occupant.

[0014] The air conditioning control method according to an embodiment of the present invention may further include a step of controlling the air conditioner based on the current temperature if the occupancy prediction probability is greater than or equal to a second threshold value when there is currently an occupant present.

[0015] The calculation step can calculate the occupancy prediction probability based on the accumulated history of indoor occupant information identified through deep learning-based object recognition in indoor images captured by a camera.

[0016] An air conditioning control system according to another embodiment of the present invention includes a processor that determines whether there is currently an occupant, calculates an occupancy prediction probability, which is a probability of occupancy after a predetermined time, increases a standby count when the occupancy prediction probability exceeds a first threshold in a state where there is currently no occupant, and initializes the standby count and turns off the air conditioner when the standby count exceeds a current maximum standby count; and a storage unit that provides storage space required by the processor.

[0017] According to another embodiment of the present invention, an air conditioning control method includes: a step of accumulating indoor occupant information identified through deep learning-based object recognition from an indoor image captured by a camera; a step of calculating an occupancy prediction probability based on a history of accumulated indoor occupant information; a step of increasing a standby count when the occupancy prediction probability exceeds a first threshold in a state where there is currently no occupant; and a step of initializing the standby count and turning off the air conditioner when the standby count exceeds a current maximum standby count.

[0018] A variable air conditioning system according to another embodiment of the present invention includes: an information collection server that accumulates and stores indoor occupant information identified through deep learning-based object recognition from indoor images captured by a camera; and an air conditioning control system that calculates an occupancy prediction probability based on a history of accumulated indoor occupant information stored in the information collection server, increases a standby count when the occupancy prediction probability exceeds a first threshold in a state where there is currently no occupant, and initializes the standby count and turns off the air conditioner when the standby count exceeds a current maximum standby count.

[0019] As described above, according to embodiments of the present invention, in addition to controlling air conditioning through occupancy probability predicted based on occupancy history, it is possible to adaptively satisfy both occupant comfort and energy saving by dynamically adjusting the current maximum standby count based on the predicted occupancy probability.

[0020] Figure 1 is a flow chart provided to explain a variable air conditioning control method according to one embodiment of the present invention;

[0021] FIG. 2 is a diagram illustrating the configuration of a variable air conditioning system according to another embodiment of the present invention;

[0022] Figure 3 is a diagram schematically showing the process of identifying and accumulating indoor occupant information.

[0023] Figure 4 is a detailed configuration diagram of the air conditioning control system.

[0024] Figure 5 is a diagram schematically showing the process of calculating the occupancy prediction probability of the air conditioning control system.

[0025] Figure 6 is a structural diagram for controlling air conditioners for each room by adjusting dampers in the air conditioning control system.

[0026] Hereinafter, the present invention will be described in more detail with reference to the drawings.

[0027] AI technology is making previously impossible feasible in a variety of fields. One such technology is camera-based room-level occupancy recognition within buildings.

[0028] This allows for the transmission and storage of digital information on the number of occupants per room, utilizing anonymized recognition data from real-time video. Accumulating this digital information generates occupancy history data by time zone and day of the week for each room. By accumulating weekly data on a monthly basis, the occupancy probability for each room can be calculated.

[0029] The probability of occupancy obtained in this way can be used to link with AI-based air conditioning control, but the problem is that in an unpredictable situation, such as when occupancy is predicted but no one comes, it is not possible to know how long to wait before turning off the air conditioning.

[0030] In an embodiment of the present invention, a variable air conditioning control method is proposed through occupancy probability prediction based on occupancy history to solve this problem.

[0031] This is a technology that improves the comfort of residents by preemptively performing air conditioning based on the prediction of occupancy probability when controlling heating and cooling, and saves energy by predicting unoccupied rooms in advance and stopping air conditioning operation in advance.

[0032] In addition, pre-conditioning is performed based on the occupancy probability prediction, but if actual occupancy is not achieved, the pre-conditioning stop point is variably suggested, thereby adaptively adjusting the comfort of residents and energy savings.

[0033] FIG. 1 is a flowchart provided to explain a variable air conditioning control method according to one embodiment of the present invention.

[0034] As illustrated, first, the presence of a current occupant in the individual room subject to air conditioning control is determined (S105), and then the predicted occupancy probability, which is the probability of occupancy after a specified time, is calculated (S110). In step S110, the "after the specified time" refers to a specific point in the future, such as 10 minutes later or 20 minutes later.

[0035] If the current occupancy is not present (S115-N), and the predicted occupancy probability is below the occupancy threshold, i.e., the likelihood of future occupancy is low (S120-N), the standby count is reset and the air conditioner is turned off (S125). Because there is currently no occupant and the likelihood of future occupancy is low, the air conditioner is not operated. The occupancy threshold can be set to, for example, 70%.

[0036] On the other hand, if there is no occupant currently (S115-N), and the occupancy prediction probability exceeds the occupancy threshold, i.e., if there is a high possibility that there will be an occupant in the future (S120-Y), the standby count is increased by one (S130).

[0037] If the standby count is less than or equal to the current maximum standby count (S135-N), it is determined whether the current temperature matches the set temperature (S140). If the current temperature is determined to match the set temperature (S140-Y), the standby count is reset and the air conditioner is turned off (S125). On the other hand, if the current temperature is determined to not match the set temperature (S140-N), the air conditioner is turned on and controlled (S145). Although there is currently no one in the room, there is a high possibility that there will be an occupant in the future, so the air conditioner is operated in advance until the set temperature is reached.

[0038] On the other hand, if the waiting count exceeds the current maximum waiting count (S135-Y), the waiting count is reset and the air conditioner is turned off (S125). The current maximum waiting count varies depending on the occupancy prediction probability calculated in step S110, and is specifically determined according to the following formula.

[0039] Current maximum waiting count = maximum waiting count × occupancy prediction probability

[0040] The maximum standby count is an arbitrarily set count value. For example, if the maximum standby count is 10 and the occupancy prediction probability is 90%, the current maximum standby count is 9. Assuming that the steps in Figure 1 operate in 1-minute cycles, if there is no occupant after 9 minutes, the air conditioner will turn off. In the same setting, if the occupancy prediction probability is 70%, the current maximum standby count is 7, and if there is no occupant after 7 minutes, the air conditioner will turn off.

[0041] This is a technology configuration that reasonably determines when to stop preemptive strikes when the probability of prediction is high and the prediction is confirmed to have failed.

[0042] Meanwhile, if there is currently an occupant (S115-Y), and the occupancy prediction probability is greater than the non-occupancy threshold, i.e., if there is a high possibility that there will continue to be an occupant in the future (S150-N), the process starts from step S140, and if the current temperature is determined to match the set temperature (S140-Y), the standby count is reset and the air conditioner is turned off (S125). On the other hand, if the current temperature is determined to not match the set temperature (S140-N), the air conditioner is turned on and controlled (S145).

[0043] On the other hand, if the predicted occupancy probability is below the non-occupancy threshold while the current occupancy is present (S115-Y), i.e., if the likelihood of continued occupancy is low (S150-Y), the standby count is reset and the air conditioner is turned off (S155). This configuration aims to save energy by predicting non-occupancy in advance and stopping air conditioning operation in advance.

[0044] Afterwards, whenever a period (e.g., 1 minute) arrives, the process is repeated from step S105.

[0045] FIG. 2 is a diagram illustrating the configuration of a variable air conditioning system according to another embodiment of the present invention. As illustrated, the variable air conditioning system according to the embodiment of the present invention is configured to include IoT sensors (210), an occupancy information collection server (220), and an air conditioning control system (230).

[0046] IoT sensors (210) identify the number of occupants in a room through deep learning-based object recognition in indoor images captured by cameras. The occupancy information collection server (220) collects and stores information on the number of occupants identified by the IoT sensors (210) for each room. Fig. 3 schematically illustrates the process of accumulating and storing information on indoor occupancy identified by the IoT sensors (210) as history in the occupancy information collection server (220).

[0047] The air conditioning control system (230) performs variable air conditioning according to the algorithm illustrated in Fig. 1 based on the occupancy information stored by the occupancy information collection server (220). Fig. 4 is a drawing illustrating a detailed configuration of the air conditioning control system (230).

[0048] According to another embodiment of the present invention, an air conditioning control system (230) can be implemented as a computing system including a communication unit (231), an output unit (232), a processor (233), an input unit (234), and a storage unit (235), as shown.

[0049] The communication unit (231) is a communication interface for connection with an external network or external device, and obtains presence history information from the presence information collection server (220). The output unit (220) is an output means for displaying the results of calculations performed by the processor (230), and the input unit (240) is a user interface for receiving user commands and transmitting them to the processor (230).

[0050] The processor (230) calculates the occupancy prediction probability according to the procedure illustrated in the aforementioned Figure 1 and performs air conditioning, pre-conditioning, etc. based on the same. The storage unit (250) provides the storage space necessary for the processor (230) to function and operate.

[0051] Figure 5 schematically illustrates the process (S110) of calculating the occupancy prediction probability of an air conditioning control system (230). The occupancy history data stored in the occupancy information collection server (220) is accumulated by time zone, day of the week, week, and month, and used as the basis for calculating the occupancy prediction probability of each room.

[0052] Figure 6 schematically illustrates a structure for controlling air conditioners for each room by adjusting dampers in an air conditioning control system (230).

[0053] So far, a preferred embodiment of a variable air conditioning control method and system using occupancy probability prediction based on occupancy history has been described in detail.

[0054] In the above example, only the current maximum waiting count was dynamically operated, but other parameters can also be dynamically operated. For example, the maximum waiting count, the presence threshold, and the absence threshold can also be dynamically operated.

[0055] Specifically, if the occupancy prediction probability does not accurately predict future actual occupancy—for example, if the occupancy prediction probability is incorrect a certain number of times over a given period—then the maximum waiting count can be decreased. Conversely, if the occupancy prediction probability accurately predicts future actual occupancy, then the maximum waiting count can be increased.

[0056] Furthermore, if the occupancy prediction probability does not accurately predict future actual occupancy—for example, if the occupancy prediction probability is incorrect a certain number of times over a given period—then the occupancy threshold can be raised and the non-occupancy threshold lowered. Conversely, if the occupancy prediction probability accurately predicts future actual occupancy, then the occupancy threshold can be lowered and the non-occupancy threshold raised.

[0057] Meanwhile, it goes without saying that the technical idea of ​​the present invention can also be applied to a computer-readable recording medium containing a computer program that performs the functions of the device and method according to the present embodiment. In addition, the technical idea according to various embodiments of the present invention can be implemented in the form of computer-readable code recorded on a computer-readable recording medium. The computer-readable recording medium can be any data storage device that can be read by a computer and store data. For example, the computer-readable recording medium can be a ROM, a RAM, a CD-ROM, a magnetic tape, a floppy disk, an optical disk, a hard disk drive, etc. In addition, the computer-readable code or program stored on the computer-readable recording medium can be transmitted through a network connected between computers.

[0058] In addition, although the preferred embodiments of the present invention have been illustrated and described above, the present invention is not limited to the specific embodiments described above, and various modifications can be made by a person having ordinary skill in the art to which the present invention pertains without departing from the gist of the present invention as claimed in the claims, and such modifications should not be understood individually from the technical idea or prospect of the present invention.

Claims

1. Step to determine whether there is currently a resident; A step of calculating the occupancy prediction probability, which is the probability of occupancy after a set time; A step of increasing the standby count when the occupancy prediction probability exceeds the first threshold while there is no occupant currently present; An air conditioning control method characterized by including a step of initializing the standby count and turning off the air conditioner when the standby count exceeds the current maximum standby count.

2. In claim 1, The current maximum waiting count is: An air conditioning control method characterized in that it varies according to the occupancy prediction probability.

3. In claim 2, The current maximum waiting count is: Current maximum waiting count = maximum waiting count × occupancy prediction probability An air conditioning control method characterized by being calculated using the above formula.

4. In claim 2, An air conditioning control method, characterized in that it further includes a step of controlling the air conditioner based on the current temperature if the standby count is less than or equal to the maximum standby count.

5. In claim 4, The air conditioner control stage is: If the standby count is less than or equal to the maximum standby count, a step of determining whether the current temperature matches the set temperature; When the current temperature matches the set temperature, the standby count is reset and the air conditioner is turned off; An air conditioning control method characterized by including a step of turning on and controlling an air conditioner if the current temperature does not match the set temperature.

6. In claim 4, An air conditioning control method characterized by further comprising a step of initializing a standby count and turning off the air conditioner when the occupancy prediction probability is lower than a first threshold value in a state where there is no occupant currently present.

7. In claim 6, An air conditioning control method characterized by further comprising a step of initializing a standby count and turning off the air conditioner when the occupancy prediction probability is lower than a second threshold value that is lower than a first threshold value in a state where there is currently an occupant.

8. In claim 7, An air conditioning control method characterized by further comprising a step of controlling an air conditioner based on a current temperature when the occupancy prediction probability is greater than or equal to a second threshold value in a state where a current occupant is present.

9. In claim 1, The calculation steps are: An air conditioning control method characterized by calculating an occupancy prediction probability based on a history of accumulated indoor occupant information identified through deep learning-based object recognition in indoor images captured by a camera.

10. A processor that determines whether there is currently an occupant, calculates the occupancy prediction probability, which is the probability of occupancy after a set time, increases the standby count when the occupancy prediction probability exceeds the first threshold while there is currently no occupant, and resets the standby count and turns off the air conditioner when the standby count exceeds the current maximum standby count; and An air conditioning control system comprising: a storage unit providing storage space required by a processor; 11. A step of accumulating indoor occupant information identified through deep learning-based object recognition from indoor images captured by a camera; A step of calculating the probability of occupancy prediction based on the accumulated history of indoor occupancy information; A step of increasing the standby count when the occupancy prediction probability exceeds the first threshold while there is no occupant currently present; and An air conditioning control method characterized by including a step of initializing the standby count and turning off the air conditioner when the standby count exceeds the current maximum standby count.

12. An information collection server that accumulates and stores indoor occupant information identified through deep learning-based object recognition from indoor images captured by a camera; and A variable air conditioning system characterized by comprising an air conditioning control system that calculates an occupancy prediction probability based on a history of accumulated indoor occupancy information stored in an information collection server, increases a standby count when the occupancy prediction probability exceeds a first threshold while there is currently no occupant, and initializes the standby count and turns off the air conditioner when the standby count exceeds the current maximum standby count.

Images