Electronic apparatus and operation method for predicting HVAC energy consumption

1. An operation method of an air conditioner, the operation method comprising:

obtaining, by the air conditioner, feature information on usage patterns of the air conditioner according to weather conditions;

transmitting, by the air conditioner, the feature information on the usage patterns of the air conditioner according to the weather conditions to a server as training data for an artificial intelligence model;

obtaining, by the air conditioner, weather forecast information;

transmitting, by the air conditioner, the weather forecast information to the server:

receiving, by the air conditioner, a predicted power consumption amount to be consumed by the air conditioner from the server, wherein the predicted power consumption amount is outputted from the artificial intelligence model when the weather forecast information is inputted to the artificial intelligence model;

displaying, by the air conditioner, the predicted power consumption amount of the first air conditioner output from the artificial intelligence model;

displaying, by the air conditioner, a notification indicating that the predicted power consumption amount of the air conditioner is within a predetermined range of a threshold value when the predicted power consumption amount of the air conditioner is within the predetermined range of the threshold value; and

performing, by the air conditioner, a power saving operation when the predicted power consumption amount of the air conditioner is within the predetermined range of the threshold value,

wherein the artificial intelligence model is trained, by the server, such that, when a weather condition of the weather conditions is inputted to the artificial intelligence model, the artificial intelligence model outputs an amount of power consumed by the air conditioner obtained from a feature information corresponding the weather condition by using the feature information on the usage patterns of the air conditioner according to the weather conditions as the training data.

2. The operation method of claim 1 , wherein the artificial intelligence model is trained to obtain correlation information, by learning weather history information, an amount of power consumed by each of a plurality of air conditioners, operation history information of each of the plurality of air conditioners, status information of each of the plurality of air conditioners, and type information of each of the plurality of air conditioners.

3. The operation method of claim 2 , wherein the artificial intelligence model is trained to:

obtain operation patterns of the plurality of air conditioners according to the weather condition, by learning weather history information of a first duration and operation history information of each of the plurality of air conditioners of the first duration, and

obtain first correlation information based on the obtained operation patterns.

4. The operation method of claim 3 , wherein the artificial intelligence model is trained to:

obtain second correlation information, based on a weather history of a second duration after the first duration, operations of the plurality of air conditioners related to the weather history of the second duration, and the first correlation information, and predict the amount of power to be consumed by the air conditioner, based on the second correlation information.

5. The operation method of claim 2 , wherein the artificial intelligence model is trained to:

predict and output a first power amount to be consumed by the air conditioner during a third duration, based on weather forecast information of the third duration and the correlation information,

predict a second power amount to be consumed by the air conditioner during a fourth duration, based on weather forecast information of the fourth duration after the third duration, the predicted first power amount, and the correlation information, and

output the second power amount.

6. The operation method of claim 1 , further comprising:

comparing, by the air conditioner, a threshold value input from a user with the predicted power consumption amount, and outputting a time at which the predicted power consumption amount is equal to or greater than the threshold value.

7. The operation method of claim 1 , further comprising:

obtaining, by the air conditioner, personal information of a user;

inputting, by the air conditioner, the personal information of the user to the artificial intelligence model; and

transmitting, by the air conditioner, a first control signal for a power saving operation procedure to be performed by the air conditioner, the first control signal being output from the artificial intelligence model to the air conditioner,

wherein the artificial intelligence model is trained to generate the first control signal based on the personal information.

8. The operation method of claim 2 , wherein the artificial intelligence model is trained to:

obtain feature information about an operation pattern of each of the plurality of air conditioners from operation history information of each of the plurality of air conditioners,

generate a first class comprising the air conditioner, by clustering the air conditioner and the plurality of air conditioners having operation patterns similar to an operation pattern of the air conditioner based on the obtained feature information,

obtain first correlation information between the weather condition and power consumption amounts of the air conditioners included in the first class, based on the weather history and pieces of feature information of the air conditioners included in the first class, and

predict the amount of power to be consumed by the air conditioner based on the first correlation information and the weather forecast information.

9. A computer program product comprising a non-transitory computer-readable storage medium having recorded thereon a plurality of instructions, which, when executed, instruct at least one processor to:

obtain, by an air conditioner, feature information on usage patterns of the air conditioner according to weather conditions,

transmit, by the air conditioner, the feature information on the usage patterns of the air conditioner according to the weather conditions to a server as training data for an artificial intelligence model,

obtain, by the air conditioner, weather forecast information,

transmit, by the air conditioner, the weather forecast information to the server,

receive, by the air conditioner, a predicted power consumption amount to be consumed by the air conditioner from the server, wherein the predicted power consumption amount is outputted from the artificial intelligence model when the weather forecast information is inputted to the artificial intelligence model,

display, by the air conditioner, the predicted power consumption amount of the air conditioner, output from the artificial intelligence model,

display, by the air conditioner, a notification indicating that the predicted power consumption amount of the air conditioner is within a predetermined range of a threshold value when the predicted power consumption amount of the air conditioner is within the predetermined range of the threshold value, and

perform, by the air conditioner, a power saving operation when the predicted power consumption amount of the air conditioner is within the predetermined range of the threshold value,

wherein the artificial intelligence model is trained, by the server, such that, when a weather condition of the weather conditions is inputted to the artificial intelligence model, the artificial intelligence model outputs an amount power consumed by the air conditioner obtained from a feature information corresponding the weather condition by using the feature information on the usage patterns of the air conditioner according to the weather conditions as the training data.

10. An air conditioner comprising:

a memory configured to store at least one instruction;

a communication circuit;

a display; and

at least one processor coupled to the memory,

wherein the at least one instruction, which when executed, configures the at least one processor to:

obtain feature information on usage patterns of the air conditioner according to weather conditions;

control the communication circuit to transmit the feature information on the usage patterns of the air conditioner according to the weather conditions to a server as training data for at least one artificial intelligence model;

obtain weather forecast information;

control the communication circuit to transmit the weather forecast information to the server;

control the communication circuit to receive a predicted power consumption amount to be consumed by the air conditioner from the server, wherein the predicted power consumption amount is outputted from the at least one artificial intelligence model when the weather forecast information is inputted to the at least one artificial intelligence model;

control the display to display the predicted power consumption amount of the air conditioner output from the at least one artificial intelligence model, and

control the display to display a notification indicating that the predicted power consumption amount of the air conditioner is within a predetermined range of a threshold value when the predicted power consumption amount of the air conditioner is within the predetermined range of the threshold value; and

perform a power saving operation when the predicted power consumption amount of the air conditioner is within the predetermined range of the threshold value,

wherein the at least one artificial intelligence model is trained, by the server, such that, when a weather condition of the weather conditions is inputted to the at least one artificial intelligence model, the at least one artificial intelligence model outputs an amount power consumed by the air conditioner obtained from a feature information corresponding the weather condition by using the feature information on the usage patterns of the air conditioner according to the weather conditions as the training data.

11. The air conditioner of claim 10 , wherein the at least one artificial intelligence model is further trained to obtain correlation information, by learning weather history information, an amount of power consumed by each of a plurality of air conditioners, operation history information of each of the plurality of air conditioners, status information of each of the plurality of air conditioners, and type information of each of the plurality of air conditioners.

12. The air conditioner of claim 11 , wherein the at least one artificial intelligence model comprises a pattern analysis model trained to:

obtain operation patterns of the plurality of air conditioners according to the weather condition, by learning weather history information of a first duration and operation history information of each of the plurality of air conditioners of the first duration, and

obtain first correlation information based on the obtained operation patterns.

13. The air conditioner of claim 12 ,

wherein the pattern analysis model is trained to obtain second correlation information, based on a weather history of a second duration after the first duration, operations of the plurality of air conditioners related to the weather history of the second duration, and the first correlation information, and

wherein the at least one artificial intelligence model further comprises a predicting model trained to predict the amount of power to be consumed by the air conditioner, based on the second correlation information.

14. The air conditioner of claim 11 , wherein the at least one artificial intelligence model comprises a predicting model trained to:

predict and output a first power amount to be consumed by the air conditioner during a third duration based on weather forecast information of the third duration and the correlation information, and

predict and output a second power amount to be consumed by the air conditioner during a fourth duration based on weather forecast information of the fourth duration after the third duration, the predicted first power amount, and the correlation information.

15. The air conditioner of claim 10 , wherein, the at least one instruction, which when executed, further configures the at least one processor to compare a threshold value input from a user with the predicted power consumption amount and output a time at which the predicted power consumption amount is equal to or greater than the threshold value.

16. The air conditioner of claim 10 , further comprising:

a communication circuit configured to transmit/receive data to/from the first air conditioner,

wherein the at least one instruction, which when executed, further configures the at least one processor to:

obtain personal information of a user;

input the personal information of the user to the at least one artificial intelligence model; and

control the communication circuit to transmit, to the air conditioner, a first control signal for a power saving operation procedure to be performed by the air conditioner, the first control signal being output from the at least one artificial intelligence model, and

wherein the at least one artificial intelligence model further comprises a control model trained to generate the first control signal based on the personal information.

17. The air conditioner of claim 11 , wherein the at least one artificial intelligence model comprises:

a pattern analysis model trained to:

obtain feature information about an operation pattern of each of a plurality of air conditioners from operation history information of each of the plurality of air conditioners,

generate a first class comprising the air conditioner by clustering air conditioners having operation patterns similar to an operation pattern of the air conditioner based on the obtained feature information, and

obtain first correlation information between the weather condition and power consumption amounts of the air conditioners included in the first class based on the weather history and pieces of feature information of the air conditioners included in the first class; and

a predicting model trained to predict the amount of power to be consumed by the air conditioner based on the first correlation information and the weather forecast information.

18. The air conditioner of claim 10 , further comprising:

a communication circuit configured to transmit/receive data to/from a server,

wherein the at least one instruction, which when executed, further configures the at least one processor to:

obtain operation history information of the air conditioner and input the operation history information to a first artificial intelligence model; and

control the communication circuit to transmit first feature information about an operation pattern of the air conditioner output from the first artificial intelligence model to the server and receive, from the server, information about the predicted power consumption amount of the air conditioner obtained by the server by using a second artificial intelligence model,

wherein the first artificial intelligence model is trained to obtain and output the first feature information based on the operation history information of the air conditioner, and

wherein the second artificial intelligence model is trained to obtain and output the information about the predicted power consumption amount of the air conditioner based on the first feature information and the weather forecast information.