RU2772237C2 - Data-based learning server and method for forming and using its learning model - Google Patents

Abstract

FIELD: air conditioning control.

SUBSTANCE: invention relates to a server, methods and equipment for data-based learning and air conditioning control. The server comprises a communication unit, a data storage unit and at least one processor configured to receive the set temperature set in the air conditioner and the current ambient temperature in the air conditioner room during the setting of the set temperature through the communication unit, apply the set temperature and the current ambient temperature environment in the training model, and obtain a recommended temperature set in the air conditioner, the training model learning to provide the recommended temperature based on the set temperature and the current ambient temperature in the room.

EFFECT: invention provides the possibility of determining and controlling the temperature of the air conditioner.

30 cl, 20 dwg

Description

The technical field to which the invention belongs

[1] The disclosure relates to a method for generating a training model and a data-based learning server using the generated training model.

State of the art

[2] In recent years, intelligent services that automatically recognize data such as voice, image, moving image, and text so as to provide data-related information or data-related services have been used in various fields of technology.

[3] The artificial intelligence technology used in intelligent services is a technology that realizes human-level intelligence. Unlike existing rule-based intelligent systems, AI technology allows machines to perform learning and judgment and become intelligent spontaneously for the machine. Since artificial intelligence technology is used, the recognition rate increases and users can be understood more accurately, so that the existing rule-based technology is gradually replaced by artificial intelligence technology.

[4] Artificial intelligence technology includes technologies based on machine learning and elements that use machine learning.

[5] Machine learning is an algorithmic technology that spontaneously classifies/trains features of input data. Element-based technology is a technology that models functions such as recognition and judgment of the human brain using machine learning algorithms, and includes fields of technology such as linguistic understanding, visual understanding, inference/prediction, knowledge representation, and control. movement.

[6] Applications of artificial intelligence technology are different as follows. Linguistic understanding is a technology for natural language/character recognition and application/processing, and includes natural language processing, machine translation, conversational system, query response, speech recognition/synthesis, and the like. Visual understanding is a technology to recognize and process objects such as human vision, and includes object recognition, object tracking, image retrieval, human recognition, scene understanding, spatial understanding and image enhancement, and the like. Inference-based prediction is a technology for judging and inferring and predicting information, and includes knowledge/probability-based inference, optimization-based prediction, preference-based planning, recommendation, and the like. Knowledge representation is a technology for automating human experience information into knowledge data and includes knowledge construction (data generation/classification), knowledge management (data use), and the like. The motion control is a technology for controlling the automatic driving of a vehicle and the motion of a robot, and includes motion control (navigation, collision, driving), functional control (behavior control), and the like.

[7] The above information is provided as background information only to assist in understanding the disclosure. No determinations are made and no judgments are made as to whether or not any of the foregoing may or may not be applicable as prior art regarding the disclosure.

The essence of the invention

Technical task

[8] Exemplary embodiments of the disclosure overcome the above disadvantages and other disadvantages not described above. Furthermore, the present invention need not overcome the disadvantages described above, and an exemplary embodiment of the disclosure may not overcome any of the problems described above.

[9] The disclosure of the essence is to set the temperature of the air conditioner using artificial intelligence technology.

The solution of the problem

[10] Accordingly, the disclosure is to provide a method for generating and using a training model for setting the temperature of an air conditioner.

[11] In addition, the technical subjects of the disclosure are not limited to the technical matters described above, and other technical subjects that are not mentioned can be clearly understood by those skilled in the art to which the disclosure relates from the description below.

[12] In accordance with the disclosure aspect, a data-based learning server is provided. The data-based learning server includes a communication unit configured to communicate with an external device, at least one processor configured to receive a set temperature set in the air conditioner and the current temperature of the air conditioner at the time of temperature setting through the communication unit, and generate or replaced with a new training model using the set temperature and the current temperature, and a data storage unit configured to store the generated or replaced with a new training model to provide the recommended temperature to be set in the air conditioner, as a result of the formation or replacement with a new training models.

[13] In accordance with another aspect of the disclosure, a data-based learning server is provided. The data-based learning server includes a data storage unit configured to store a trained training model to provide a recommended temperature to be set to an air conditioner, at least one processor configured to obtain the current temperature of the air conditioner and input the current temperature to the air conditioner. a training model to obtain a recommended temperature to be set in the air conditioner; and a communication unit configured to transmit the recommended temperature to an external device.

[14] In accordance with another aspect of the disclosure, a network system is provided.

The system includes an air conditioner and a training model server configured to generate a training model using training data received from the air conditioner, wherein the air conditioner includes a temperature sensor configured to read the current temperature near the air conditioner, a fan configured to emit cooling air outside based on the set temperature set in the air conditioner, and an air conditioner communication unit configured to communicate with an external device, and at least one air conditioner processor configured to control the air conditioner communication unit so as to transmit the read current temperature and the set temperature to an external device, and the training model server includes at least one server processor that obtains the current temperature and the set temperature, and generates a training model using the obtained set temperature and the current th temperature, and a data storage unit configured to store the generated training model to provide the recommended temperature of the air conditioner as a result of generating the training model.

[15] In accordance with another aspect of the disclosure, a network system is provided.

The network system includes an air conditioner and a training model server configured to provide a recommended temperature using recognition data obtained from the air conditioner, the air conditioner includes a temperature sensor configured to read the current temperature of the air conditioner, a fan configured to issue a cooling air generated from the air cleaner to the outside, and an air conditioner communication unit transmitting the current temperature to the first external device, wherein the training model server includes a data storage unit configured to store the trained training model to provide the recommended temperature of the air conditioner, at least one a server processor configured to obtain the current temperature and input the current temperature to the training model to obtain the recommended temperature of the air conditioner, and a server communication unit configured to transmit the recommended temperature to the second external device.

[16] In accordance with another aspect of the disclosure, an air conditioner is provided.

The air conditioner includes a fan configured to discharge cooling air to the outside, a temperature sensor configured to read the current temperature near the air conditioner, a communication unit configured to exchange data with an external device, and at least one processor configured to control the unit connection so as to transmit the current temperature to an external device, control the communication unit so as to receive the recommended temperature, which is the result obtained by applying the current temperature to the training model, from the external device depending on the transmission of the current temperature, and set the received a recommended temperature in the air conditioner, wherein the training model is a training model trained using a plurality of target temperatures preset in the air conditioner and a plurality of current temperatures.

[17] In accordance with another aspect of the disclosure, a user terminal is provided. The user terminal includes a display configured to display a screen, a communication unit configured to communicate with an external device, an input configured to receive user input, and at least one processor configured to control the communication unit in a manner to transmit the artificial intelligence operation request signal corresponding to the artificial intelligence operation UI to the air conditioner in response to the user input signal dependent on the user input selecting the artificial intelligence operation UI included in the screen which is received via the input, and control the display so as to display the received recommended temperature in response to receiving the recommended temperature set in the air conditioner, which is the result obtained by applying the current temperature of the air conditioner to the training model depending on the artificial operation request signal intelligence through the communication unit.

[18] In accordance with another aspect of the disclosure, a method is provided for generating a training model for a training server based on data. The method includes obtaining the set temperature set in the air conditioner and the current temperature of the air conditioner at the time of setting the temperature, generating or replacing with a new training model using the set temperature and the current temperature, and storing the generated or replaced with the new training model to provide the recommended temperature, which should be set in the air conditioner as a result of formation or replacement with a new training model.

[19] In accordance with another aspect of the disclosure, a method is provided for using a training model for a data-based learning server. The method includes storing the trained training model to provide a recommended temperature to be set to the air conditioner, obtaining the current temperature of the air conditioner, inputting the current temperature to the trained training model to obtain the recommended temperature to be set to the air conditioner, and passing the recommended temperature to the air conditioner. external device.

[20] In accordance with another aspect of the disclosure, a method is provided for providing a recommended air conditioner temperature. The method includes reading the current temperature of the air conditioner, transmitting the read current temperature to an external device, receiving the recommended temperature, which is the result obtained by applying the current temperature to the training model, from the external device depending on the transmission of the current temperature, and setting the received recommended temperature in the air conditioner, wherein the training model is a training model trained using the set temperature preset in the air conditioner and the current temperature set.

[21] In accordance with another aspect of the disclosure, a method for controlling an air supply controller of a user terminal is provided. The method includes receiving a user input signal dependent on a user input selecting an AI operation UI, transmitting an AI operation request signal corresponding to the AI operation UI to an air conditioner, obtaining a recommended temperature set in the air conditioner, which is a result obtained by applying the current temperature of the air conditioner to the training model depending on the request signal of the artificial intelligence operation, and displaying the obtained recommended temperature on the screen.

[22] In accordance with another aspect of the disclosure, a method is provided for generating a training model of a network system including an air conditioner and a training model server. The method includes receiving, by the air conditioner, a user control signal adjusting the temperature, an operation of transmitting, by the air conditioner, the set temperature and the current temperature of the air conditioner to an external device, generating, by the training model server, a training model using the set temperature and the current temperature, and storing, by the training model server, the generated training model to provide the recommended air conditioner temperature.

[23] In accordance with another aspect of the disclosure, a method is provided for providing a recommended temperature in a network system including an air conditioner and a training model server. The method includes transmitting, by means of the air conditioner, the current temperature of the air conditioner to an external device, obtaining, via the training model server, the recommended temperature of the air conditioner by applying the current temperature to the training model, and transmitting, by means of the air conditioner, the recommended temperature to the external device.

[24] In accordance with another aspect of the disclosure, there is provided a method for controlling an air conditioner of a network system including an air conditioner and a user terminal. The method includes receiving, by the user terminal, a user input signal dependent on the user input selecting an AI operation UI, transmitting, by the user terminal, an AI operation request signal corresponding to the AI operation UI to an air conditioner, transmitting, by the air conditioner, the current temperature of the air conditioner to the external device, if the artificial intelligence operation request signal is received, receiving, by the air conditioner, the recommended temperature, which is the result obtained by applying the current temperature to the training model, from the external device depending on the transmission of the current temperature, and setting, by the air conditioner receiving the recommended temperature in the air conditioner, wherein the training model is a training model trained using a plurality of preset temperatures predetermined s in the air conditioner, and a set of current temperatures.

[25] According to the embodiment of the disclosure, since the temperature set in the air conditioner is automatically recommended using artificial intelligence technology, the convenience of the user who controls the temperature can be greatly improved. In particular, it is possible to provide the user with the most ideal recommended temperature for the user.

[26] Further, according to the method of using the entity disclosure training model, the training model can be continuously updated based on the user's temperature setting history that specifies the air conditioner, and the performance of the training model can be improved, so that when the entity disclosure training model is used, the most ideal recommended temperature may be provided to the user.

[27] In this way, a training model customized for each of the users using the air conditioner can be formed, and thus an optimal recommended temperature suitable for each of the plurality of users can be provided.

[28] Additionally, the benefits that may be obtained or expected by various embodiments of the disclosure should be directly or implicitly disclosed in the detailed description of the disclosure. For example, the various benefits that can be expected through various embodiments of the disclosure should be disclosed in the detailed description, which is described below.

[29] Other aspects, advantages, and features of the disclosure will become apparent to those skilled in the art from the following detailed description, which, when considered in conjunction with the accompanying drawings, discloses various embodiments of the disclosure.

Brief description of the drawings

[30] The above and/or other aspects of the disclosure should be made more apparent by describing certain embodiments of the disclosure with reference to the accompanying drawings, in which:

[31] FIG. 1A and 1B are diagrams showing a network system for generating and using a training model according to an embodiment of the disclosure;

[32] FIG. 2A and 2B are diagrams showing a configuration of a data-based learning server according to an embodiment of the disclosure;

[33] FIG. 3A and 3B are flowcharts of a method in a network system according to an embodiment of the disclosure;

[34] FIG. 4 is a table showing an example of generating a training model according to an embodiment of the disclosure;

[35] FIG. 5 is a diagram showing an example of applying a weighting factor to training data according to an embodiment of the disclosure;

[36] FIG. 6 is a diagram showing the structure of a cloud server according to an embodiment of the disclosure;

[37] FIG. 7A and 7B are diagrams showing a training model generation procedure according to an embodiment of the disclosure;

[38] FIG. 8 is a block diagram showing a configuration of an air conditioner according to an embodiment of the disclosure;

[39] FIG. 9 is a block diagram showing a configuration of a user terminal U according to an embodiment of the disclosure;

[40] FIG. 10A and 10B are diagrams showing a user terminal screen displaying a recommended temperature according to an embodiment of the disclosure;

[41] FIG. 11 is a flowchart showing a method for generating a training model for a learning server based on data according to an embodiment of the disclosure;

[42] FIG. 12 is a flowchart showing a method of using a training model for a data-based learning server according to an embodiment of the disclosure;

[43] FIG. 13 is a flowchart showing a method for providing the recommended temperature of an air conditioner according to an embodiment of the disclosure;

[44] FIG. 14 is a flowchart showing a method for controlling an air conditioner of a user terminal according to an embodiment of the disclosure; and

[45] FIG. 15 is a flowchart of a method in a network system including a user terminal and an air conditioner according to an embodiment of the disclosure.

[46] Throughout the drawings, like reference numerals are to be understood as referring to like parts, components, and structures.

Optimal Mode for Carrying Out the Invention

[47] The following description with reference to the accompanying drawings is provided to assist in a full understanding of the various embodiments of the disclosure as defined by the claims and their equivalents. It includes various details to assist in this understanding, but they are to be regarded as merely exemplary. Accordingly, those skilled in the art will recognize that various changes and modifications to the various embodiments described herein may be made without departing from the spirit and scope of the disclosure. In addition, descriptions of well-known functions and constructions may be omitted for the sake of clarity and brevity.

[48] The disclosure of an entity based on an exemplary method is described above. The terms and words used herein are for description and are not limited to bibliographic meanings, but are used by the inventor simply to provide a clear and consistent understanding of the disclosure. The entity disclosure may be modified and changed in various ways according to the above content. Therefore, unless otherwise indicated, the disclosure is free to practice within the scope of the claims.

[49] The various embodiments set forth in the specification and the configurations shown in the drawings are merely preferred examples of the disclosures disclosed, and various modifications that may replace the various embodiments and drawings of the present specification may be present at the time of filing of this specification. applications.

[50] In addition, like references or symbols with the numbers of each drawing of the present description of the invention designate parts or components that perform almost identical functions.

[51] In addition, the terms used in the present description of the invention are used only to describe a specific embodiment, instead of limiting the disclosed disclosure of the essence. The singular forms used herein are intended to include the plural forms unless the context clearly indicates otherwise. In this detailed description, it is to be understood that the term "comprise" and variations thereof such as "comprising" and "comprises" indicate the presence of the features, numbers, steps, operations, components, parts, or combinations of the foregoing described in this detailed description, but do not exclude the presence or addition of one or more other features, numbers, steps, operations, components, parts, or combinations of the foregoing.

[52] In addition, terms including ordinal numbers such as "first" and "second" used herein may be used to describe various components, but the components are not limited by terms, and the terms are used only in order to distinguish one component from other components. For example, the "first" component may be referred to as the "second" component, and the "second" component may likewise be referred to as the "first" component, without departing from the scope of the disclosure. The term "and/or" includes a combination of a plurality of items or any of a plurality of terms.

[53] In addition, if any (eg: the first, component is "(functionally or operationally) connected (connected)" or "coupled)" with another (eg: second) component, any component can directly connect to another component or can connect to another component through another component (for example: a third component).

[54] The following describes in detail various embodiments of the disclosure with reference to the accompanying drawings.

[55] FIG. 1A and 1B are diagrams showing a network system for generating and using a training model according to an embodiment of the disclosure.

[56] Referring to FIG. 1A, the network system may include an air conditioner A (Aa or Ab), a user terminal U (Ua or Ub), and a cloud server C. Air conditioner A may be a temperature or humidity control device for indoor environments. Air conditioner A can be divided into wall-mounted, for example, air conditioner Aa, and standing, for example, air conditioner Ab.

[57] The user terminal U may be a device for controlling the air conditioner A remotely. Like the user terminal Ua, the user terminal U may be a smartphone, a cellular phone, or a tablet PC in which an application (or application) for controlling the air conditioner is installed. Alternatively, as the user terminal Ub, the user terminal U may be a remote controller (or remote controller) dedicated to the air conditioner. In addition, the user terminal U may be a smart TV, a digital camera, a personal digital assistant (PDA), a portable media player (PMP), a laptop computer, a desktop computer, and the like, but is not limited to the above examples.

[58] The user terminal U can control the air conditioner A remotely. For example, the user terminal U may use RF communication technologies such as ZigBee, Wi-Fi, Bluetooth technology, mobile communications, local area network (LAN), wide area network (WAN), Infrared Data Association (IrDA) ), UHF and VHF to send the control command to air conditioner A.

[59] The cloud server C may connect or directly connect to the air conditioner A through a third device (eg, access point (AP), repeater, router, gateway, hub, etc.).

[60] Cloud server C may include one or more servers. For example, the cloud server C may include at least one of a bridge server BS, an intelligent home service server SS, and a data-based learning server DS. In this case, two or more of the bridge server BS, the intelligent home service server SS, and the data-based learning server DS can be integrated into one server. Alternatively, at least one of the bridge server BS, the intelligent home service server SS, and the data-based learning server DS may be divided into a plurality of subservers.

[61] The bridge server BS (or device state information import server) can import state information of smart home appliances (eg, air conditioner, washing machine, refrigerator, vacuum cleaner, oven, etc.).

[62] The bridge server BS may include a connection API BS1 and a device state data DB BS2.

[63] API BS1 connections may include an application programming interface (hereinafter referred to as "API"), which serves as an interface between various devices operating depending on heterogeneous protocols. An API can be defined as a set of nested procedures or functions that can be called from any protocol for any other protocol processing. Thus, an API can provide an environment in which another protocol's work can be done in any of the protocols.

[64] The bridge server BS can import air conditioner state information using the connection API BS1. Then, the bridge server BS may store the imported air conditioner state information in the device state data DB BS2.

[65] The intelligent home service server SS (or the external environment information import server) can import the external environment information. The outside environment information may include, for example, at least one of outside temperature and outside humidity as weather information provided by the outside content server CP (eg, weather station server, weather prediction server, or the like).

[66] The data-based learning server DS may generate a training model and obtain the results of applying the training model using the trained forming model.

[67] The data-based learning server DS includes a data import API DS1, a data analytics engine DS2, an analytics DB (DS3), and an API (DS4) for providing data services.

[68] FIG. 1A shows a network system in which the learning server DS generates a learning model based on the data, and FIG. 1B shows a network system using the generated training model of the data-based learning server DS.

[69] First, the following describes the procedure of the network system in which the learning server DS generates a learning model based on the data, with reference to FIG. 1A.

[70] In step 1, air conditioner A may transmit state information (e.g., set temperature, current temperature, etc.) of air conditioner A to cloud server C via a third device (e.g., AP, repeater, router, gateway, hub etc.). The bridge server BS of the cloud server C can import the state information of the air conditioner A transmitted from the air conditioner A using the connection API BS1, and store the imported state information of the air conditioner A in the device state data DB BS2.

[71] The state information of the air conditioner A may include the set temperature set in the air conditioner A and the current temperature (eg, room temperature and ambient temperature) of the air conditioner at the time of temperature setting, depending on the user's desired temperature.

[72] The desired temperature of the user may generally be identical to the set temperature set in the air conditioner A, but may be the set temperature stepwise set by the air conditioner A until the desired temperature is reached.

[73] In addition, the current temperature (or room temperature and ambient temperature) at the time of temperature setting may include, for example, at least one of the temperature read by air conditioner A at the time of temperature setting (for example, at the time when temperature setting user operation of air conditioner A is in progress), temperatures read by air conditioner A within a certain time (for example, 10 minutes) after temperature setting, and the recent temperature read in advance before temperature setting and stored.

[74] The state information of air conditioner A may include operating mode information set in air conditioner A. The operation mode may include, for example, intelligent comfort mode, tropical sound night sleep mode, tropical sound night sleep mode in calm mode, two-stage cooling and the like, but not limited to the above mode.

[75] According to various embodiments, time information at the time of temperature setting of the air conditioner A may also be stored in the device status data DB BS2. The time information at the time of setting the temperature includes, for example, at least one of the operating time of the user who sets the temperature, the time when the bridge server BS receives the set temperature, and the time when the set temperature is stored in the device status data DB BS2.

[76] According to various embodiments, the position information of the air conditioner A may also be stored in the device state data DB BS2. In this case, the position information of air conditioner A may be stored by reception at the time of temperature setting, or stored in advance.

[77] In step 2, the intelligent home service server SS may import external environment information (or weather information) every predetermined period (for example, between 5 minutes and 30 minutes) from the communicating external content server CP and store the imported external environment information. in DB SS1 weather data.

[78] The environmental information may include at least one of outdoor temperature, outdoor humidity, dust concentration, rainfall, and sunlight amount, but is not limited to the above-described example.

[79] In steps 3 and 3', the data-learning server DS may use the data import API DS1 to obtain state information of the air conditioner A stored in the device state data DB BS2 of the bridge server BS. In addition, the data-based learning server DS may use the data import API DS1 to obtain the external environment information stored in the weather data DB SS1 of the smart home service server SS.

[80] In this case, the outside environment information is the outside information at the time of setting the temperature of the air conditioner A, and may be information retrieved from the weather data DB SS1 based on the time information at the time of setting the temperature of the air conditioner A stored in the data DB BS2 device status.

[81] Specifically, the ambient information at the time of setting the temperature may include, for example, at least one of the ambient information at the time the user sets the temperature, the ambient information in the time zone (e.g., morning/afternoon/ evening or morning/day) in which the user sets the temperature, and information of the external environment in the month or time of the year when the user sets the temperature.

[82] In addition, the ambient information may be weather information obtained based on the position information of the air conditioner A. For example, the ambient information may be weather information retrieved from the weather data DB SS1 based on the position information of the air conditioner A stored in DB BS2 device status data.

[83] In step 4, the data analytics engine DS2 of the data-based learning server DS may generate a training model using the obtained state information of the air conditioner A and the external environment information as training data.

[84] According to various embodiments, the data analytics engine DS2 of the data-based learning server DS can also generate a learning model using time information at the time of temperature setting of air conditioner A as training data.

[85] In addition, the data analytics engine DS2 can also generate a plurality of training models for each operating mode of air conditioner A at the time of temperature setting of air conditioner A.

[86] For example, the data analytics engine DS2 may generate a training model available in a smart comfort mode, a training model available in a tropical sound sleep mode, a training model available in a tropical sound sleep mode in a calm mode, and a training model available in a tropical sound sleep mode in a calm mode, and a training model available in the two-stage cooling mode, respectively.

[87] In addition, the data-based learning server DS may be executed, for example, in units of time, day, and month, as a simulation period during which the data-based learning server DS generates a training model (or updates a training model) using training data, or may be executed at the time of generating an event, but the simulation period is not limited to the above period.

[88] The following describes in more detail a process for the data-based learning server DS to generate a training model with reference to FIG. 4, 5 and 7.

[89] In step 5, the data-based learning server DS may store the generated training model in the analytic DB DS3. In this case, the training model may not be a general training model, but may be a training model configured or built to provide the recommended temperature of air conditioner A.

[90] Referring to FIG. 1B, the following describes the procedure of the network system using the training model generated by the data-based learning server DS.

[91] In step 6, the air conditioner A may receive a control command requesting execution (for example, turning on the AI mode) of the artificial intelligence function from the user terminal U.

[92] In step 7, air conditioner A may transmit state information (eg, current temperature, operating mode, etc.) of air conditioner A to cloud server C via a third device (eg, access point AP). The learning server DS based on the cloud server C can obtain state information of the air conditioner A using the DS4 API to provide data services.

[93] In step 8, the data-based learning server DS may input the obtained state information of the air conditioner A as a trained training model to provide the recommended temperature of the air conditioner A stored in the analysis DB DS3.

[94] In step 9 , the data-based learning server DS can obtain the recommended temperature of the air conditioner A as a result of applying the learning model.

[95] In step 10, the data learning server DS may transmit the obtained recommended temperature of the air conditioner A to the air conditioner A via a third device (eg, an access point AP). Further, in step 10', the data learning server DS may transmit the obtained recommended temperature of the air conditioner A to the user terminal U.

[96] In step 11, the air conditioner A that receives the recommended temperature may set the temperature of the air conditioner A as the received recommendation temperature.

[97] In addition, in step 12 , the user terminal U receiving the recommended temperature may display the received recommended temperature so that the user can confirm the received recommended temperature. Alternatively, as indicated in step 12', the user terminal U receiving the recommended temperature may display visual information indicating that the preferred recommended temperature is being adjusted compared to the preset temperature history predetermined by the user.

[98] FIG. 2A and 2B are diagrams showing a configuration of a data-based learning server according to an embodiment of the disclosure.

[99] The learning server DS based on the data of FIG. 2A is a functional block diagram for generating a training model, and the training server DS based on the data of FIG. 2B is a functional block diagram using the generated training model.

[100] FIG. 2A and 2B, the data-based learning server DS may include a communication unit 201, a data storage unit 202, and a processor 203.

[101] The communication unit 201 may communicate with an external device.

[102] The external device may include at least one of an external server (for example, a bridge server, an intelligent home service server, etc.) and an air conditioner A.

[103] The communication unit 201 may communicate with an external device in a wired or wireless communication method. Wireless communications may include, for example, cellular communications, short range communications, or Global Navigation Satellite System (GNSS) communications. Cellular communications may include, for example, Long Term Evolution (LTE), LTE-Advanced (LTE-A), Code Division Multiple Access (CDMA), Wideband CDMA (WCDMA), Universal Mobile Telecommunications System (UMTS), wireless broadband (WiBro), global system for mobile communications (GSM), etc. Near field communications may include, for example, high fidelity wireless (Wi-Fi) standard, Wi-Fi Direct, high fidelity visible light (Li-Fi), Bluetooth technology, Bluetooth Low Energy (BLE ), ZigBee, near field communication (NFC), magnetic secure data transmission, radio frequency (RF–) communication and human body area network (BAN). The communication unit 201 may also be referred to as a communication unit.

[104] The data-based learning server DS may include a data storage unit 202 . The data storage unit 202 may store the training model generated by the data-based training server DS.

[105] The storage unit 202 may include volatile and/or non-volatile storage. The volatile memory may include, for example, random access memory (RAM) (eg, DRAM, SRAM, or SDRAM). Non-volatile memory may include, for example, one-time programmable read-only memory (OTPROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), masked ROM, flash ROM, flash memory, hard disk drive, or solid state drive (SSD).

[106] The processor 203 may include one or more of a central processing unit, an application processor, a graphics processing unit (GPU), a camera image signal processor, and a communications processor (CP). According to an embodiment, the processor 203 may be implemented as a system on a chip (SoC) or a system in one package (SiP). Processor 203 may direct, for example, an operating system or application program to control at least one other component (eg, hardware or software component) of a data-based learning server (DS) coupled to processor 203 and may perform various processing and operations. with data. The processor 203 may load an instruction or data received from other components (eg, communication unit 201) into a volatile storage device, and process the downloaded instruction or data, and may store the resulting data into a non-volatile storage device.

[107] FIG. 2A and 2B are diagrams showing a configuration of a data-based learning server according to an embodiment of the disclosure.

[108] The learning server DS based on the data of FIG. 2A is a functional block diagram for generating a training model, and the training server DS based on the data of FIG. 2B is a functional block diagram using the generated training model.

[109] Referring to FIG. 2A and 2B, the data-based learning server DS may include a communication unit 201, a data storage unit 202, and a processor 203.

[110] The communication unit 201 may communicate with an external device.

[111] The external device may include at least one of an external server (for example, a bridge server, an intelligent home service server, etc.) and an air conditioner A.

[112] The communication unit 201 may communicate with an external device in a wired or wireless communication method. Wireless communications may include, for example, cellular communications, short range communications, or Global Navigation Satellite System (GNSS) communications. The cellular unit may include, for example, Long Term Evolution (LTE), LTE-Advanced (LTE-A), Code Division Multiple Access (CDMA), Wideband CDMA (WCDMA), Universal Mobile Telecommunications System (UMTS), wireless broadband (WiBro), global system for mobile communications (GSM), and the like. Near field communications may include, for example, high fidelity wireless (Wi-Fi) standard, Wi-Fi Direct, high fidelity visible light (Li-Fi), Bluetooth technology, Bluetooth Low Energy (BLE ), ZigBee, near field communication (NFC), magnetic secure data transmission, radio frequency (RF–) communication and human body area network (BAN).

[113] The data-based learning server DS may include a data storage unit 202 . The data storage unit 202 may store the training model generated by the data-based training server DS.

[114] The data storage unit 202 may include a volatile or non-volatile storage device. The volatile memory may include, for example, random access memory (RAM) (eg, DRAM, SRAM, or SDRAM). Non-volatile memory may include, for example, one-time programmable read-only memory (OTPROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), masked ROM, flash ROM, flash memory, hard disk drive, or solid state drive (SSD).

[115] The processor 203 may include one or more of a central processing unit, an application processor, a graphics processing unit (GPU), a camera image signal processor, and a communications processor (CP). According to an embodiment, the processor 203 may be implemented as a system on a chip (SoC) or a system in one package (SiP). Processor 203 may direct, for example, an operating system or application program to control at least one other component (eg, hardware or software component) of a data-based learning server (DS) coupled to processor 203 and may perform various processing and operations. with data. The processor 203 may load an instruction or data received from other components (eg, communication unit 201) into a volatile storage device, and process the downloaded instruction or data, and may store the resulting data into a non-volatile storage device.

[116] The processor 203 of FIG. 2A may be described as a functional block diagram for generating a training model.

[117] FIG. 2A, the processor 203 may include a training data acquisition block 203a and a model-based learning block 203b.

[118] The training data acquisition unit 203a may acquire the set temperature set in the air conditioner A and the current temperature of the air conditioner A at the time of temperature setting through the communication unit 201 . For example, the training data acquisition unit 203a may obtain the set temperature and the current temperature from the bridge server BS operatively connected to the air conditioner A. Alternatively, the training data acquisition unit 203a may also obtain the set temperature and current temperature from the air conditioner A or a third air conditioner a.

[119] In addition, the training data acquisition unit 203a can additionally acquire external environment information via the communication unit 201 . The outside environment information may include at least one of outside temperature and outside humidity. For example, the training data acquisition unit 203a can acquire external environment information from the intelligent home service server SS operably connected to the external content provision server CP.

[120] The model learning unit 203b may generate or update a training model using the obtained target temperature and the current temperature. When the training data acquisition unit 203a further acquires the external environment information, the model-based learning unit 203b may generate or update a training model using the set temperature, the current temperature, and the external environment information. In addition, when the training data acquisition unit 203a further acquires time information at the time of temperature setting of the air conditioner A, the model learning unit 203b can generate or update a training model using the set temperature, the current temperature, and the time information.

[121] The storage unit 202 may store the trained training model to provide a recommended temperature to be set in the air conditioner as a result of generating or updating the training model.

[122] On the other hand, when the model learning unit 203b generates or updates a plurality of training models for each operating mode of the air conditioner A, the data storage unit 202 can store the plurality of training models, respectively.

[123] The processor 203 of FIG. 2B can be described as a functional block diagram for using the training model.

[124] FIG. 2B, the processor 203 may include a recognition data acquisition unit 203c and a model application unit 203d. In this case, the storage unit 202 may store the trained training model to provide the recommended temperature to be set in air conditioner A.

[125] FIG. 2B, the recognition data acquisition unit 203c may acquire the current temperature of the air conditioner A.

[126] The model application unit 203d may input the obtained current temperature into the training model of the data storage unit 202 and obtain the recommended temperature to be set in the air conditioner A.

[127] When the recognition data acquisition unit 203c further obtains the ambient information, the model applying unit 203d may input the current temperature and the ambient information into the training model to obtain the recommended temperature to be set in the air conditioner A.

[128] In addition, when the data storage unit 202 stores a plurality of training models for each operating mode of the air conditioner A, the model applying unit 203d can set the current temperature for the training model corresponding to the current operating mode of the air conditioner A to obtain the resumed temperature of the air conditioner A.

[129] The communication unit 201 may transmit the received recommended temperature to an external device. The external device, for example, may be air conditioner A or a third device operatively connected to air conditioner A.

[130] FIG. 3A and 3B are flowcharts of a method in a network system according to an embodiment of the disclosure.

[131] A flowchart in a network system shows a data flow exchange procedure between an air conditioner A, a user terminal U, and a cloud server C.

[132] Referring to FIG. 3A and 3B, a flowchart in a network system may include a data import procedure 351 for importing training data, a data model generation procedure 352 based on training data, an artificial intelligence function control procedure 353, and a procedure 354 for setting preferred modes for each functions.

[133] FIG. 3A, air conditioner A may include a microcomputer 301 and a near field communication module 302 (eg, a Wi-Fi module). The microcomputer 301 corresponds to the processor 203 of FIG. 2A and 2B, and the near field communication module 302 may correspond to the communication unit 201 of FIG. 2A and 2B. The air conditioner A can communicate with the user terminal U and the cloud server C via the network using the near field communication module 302 . In addition, the air conditioner A can receive the recommended temperature recommended by the cloud server C through an API call associated with the near field communication module 302, and set the temperature of the air conditioner A depending on the recommended temperature.

[134] The user terminal U may include a mobile application 303 (or a mobile application). The mobile application 303 can set the artificial intelligence function and the operation mode of the air conditioner A, and perform the function of displaying the recommended temperature provided by the cloud server C on the user terminal U.

[135] The cloud server C may include a bridge server BS, a DB server (304), and a data-based learning server DS. The DB server 304 may be part of the bridge server BS or a third server physically separate from the bridge server BS.

[136] First, at step 311, the user terminal U may receive user input to change (or set) the desired temperature through the mobile application 303. Mobile application 303, for example, may be an application providing a user interface for controlling air conditioner A .

[137] At step 312, depending on the user's input, the user terminal U may send a control command to the microcomputer 301 via the near field communication module 302 to adjust the air conditioner A to the desired temperature.

[138] Alternatively, at step 311', the user can change the desired temperature via the remote control device Ub. In step 312', the remote controller Ub may transmit a control command for setting the air conditioner A to a desired temperature to the microcomputer 301 according to the user's change input.

[139] In step 313, the microcomputer 301 of the air conditioner A may generate a desired temperature change event in response to the user's desired temperature change request, and transmit the generated desired temperature change event to the bridge server BS via the near field communication unit 302. At this time, the desired temperature change event may include event data. The event data may include, for example, state information of the air conditioner A. The state information of the air conditioner A may include the desired temperature (or set temperature) and the current temperature at the time of setting the desired temperature.

[140] In addition, the event data may include, for example, operating mode information and time information of the air conditioner A. The operating mode information may include, for example, information indicating the operating mode of the air conditioner A at the time of receiving the control command of the user, or the operating mode air conditioner A when the desired temperature change event is generated. The time information may include, for example, information regarding the time when the user receives the control command, or information regarding the time when the desired temperature change event is generated.

[141] In step 314, the bridge server BS may transmit event data to the DB server 304. In step 315, the DB server 304 may store the received event data.

[142] In step 316, the DB server 304 may transmit the stored event data to the training server DS based on the data at regular periods. For example, DB server 304 may send event data daily in the form of a daily batch file. At this time, the daily batch file may include a plurality of event data. For example, when a request to change the user's desired temperature is generated several times a day, a plurality of event data may be generated, which in turn is stored in the DB server 304. The plurality of generated event data may be transmitted to the data-based learning server DS by being included in the daily batch file.

[143] In step 317, the data-based learning server DS may generate a training model using the received event data as training data. For example, the data learning server DS may generate a learning model using at least one of the set temperature of the air conditioner A, the current temperature, the environment information, the operating mode information, and the time information.

[144] In a situation in which the training model is generated as indicated in step 318, the user terminal U may receive user input that includes the artificial intelligence function of the air conditioner A. Partial screen 318a shows a portion of the screen of the user terminal U including a user interface for enable the artificial intelligence function. On the partial screen 318a, the user terminal U may receive a user input that selects a "AI custom operation" execution object 318b to enable the AI function.

[145] In step 319, depending on the user input, the user terminal U may send an AI function activation command to the microcomputer 301 via the near field communication unit 302 to turn on the AI function of the air conditioner A.

[146] Based on the AI function activation command, the microcomputer 301 may transmit device state information indicating that the AI function of the air conditioner A is being activated to the user terminal U via the near field communication module 302, as indicated at step 320. In this case, the state information of the air conditioner A can be transmitted by including in the notification event.

[147] Alternatively, as indicated in step 321, the user terminal U may transmit a device information request command requesting state information of the air conditioner A to the microcomputer 301 via the near field communication module 302 . The device information request command may be transmitted by including, for example, in a GET DEVICE message.

[148] Based on the device information request command, the microcomputer 301 may send a device information response to the user terminal U via the near field communication module 302, as indicated in step 322. In this case, the device information response may include information an artificial intelligence setting indicating that the artificial intelligence function of the air conditioner A is set to be turned on, as the state information of the air conditioner A.

[149] Thus, by considering a situation where a plurality of user terminals U are provided to control the artificial intelligence function of the air conditioner A, the air conditioner A can notify the user terminal U as to whether the artificial intelligence function of the air conditioner A is activated periodically or when an event is generated.

[150] In this regard, when the artificial intelligence function of the air conditioner A is activated, the user terminal U may receive user input to set the operating mode.

[151] Referring to FIG. 3B, at step 323, the user terminal U may receive a user input requesting execution of the intelligent comfort mode.

[152] At step 324, depending on the input of the user, the user terminal U may transmit a smart comfort control command to the microcomputer 301 via the near field communication unit 302 to perform the smart comfort mode of the air conditioner A.

[153] Based on the intelligent comfort control command, the microcomputer 301 may transmit a request command for the recommended temperature (or preferred temperature) to the data-based learning server DS via the near field communication module 302, as indicated in step 325. At this time, the command The recommended temperature request command may include, for example, the current temperature of the air conditioner A as state information of the air conditioner A. Alternatively, the recommended temperature request command may further include at least one of operating mode information indicating the current operating mode and position air conditioner information A.

[154] In step 326, the data-based learning server DS may obtain the recommended temperature of the air conditioner A as a result of applying the training model of the air conditioner A state information. data-based learning server DS to obtain the recommended temperature of air conditioner A.

[155] In this case, the data-based learning server DS may apply the state information of the air conditioner A to the training model corresponding to the operating mode of the air conditioner A based on the operating mode information of the air conditioner A to obtain the recommended temperature of the air conditioner A. In the embodiment, the server DS data-based learning can obtain the recommended temperature of the air conditioner A by applying the status information of the air conditioner A to the training model corresponding to the intelligent comfort mode.

[156] After the recommended temperature is obtained, in step 327, the data-based learning server DS may transmit the obtained recommended temperature to the microcomputer 301 via the near field communication module 302.

[157] At step 328, the microcomputer 301 receiving the recommended temperature may change the recommended temperature to the set temperature. Then, the microcomputer 301 may control the air conditioner A depending on the changed set temperature.

[158] On the other hand, if there is no response from the data-based learning server DS within a predetermined time (for example, 30 seconds) 329 after the microcomputer 301 requests the recommended temperature to the data-based learning server DS, at step 330, the microcomputer 301 can maintain the existing set temperature. The existing setpoint temperature may, for example, be a predetermined temperature before a user input for requesting execution of the smart comfort mode, a predetermined temperature corresponding to the current operating mode (eg, smart comfort mode), and the like.

[159] FIG. 4 is a table showing an example of generating a training model according to an embodiment of the disclosure.

[160] Referring to FIG. 4, the data-based learning server DS may perform a learning procedure 404 using different training data 403 depending on the air conditioner A type 401 and the air conditioner A mode 402. For example, the air conditioner A type 401 may include a floor standing air conditioner (FAC) type (or standing air conditioner) and type of room air conditioner (RAC) (wall air conditioner). In this case, the data-based learning server DS can generate training models corresponding to each of intelligent comfort mode, tropical sound night sleep mode, and tropical sound night sleep mode in calm mode as the operation mode of the floor air conditioner. In addition, the data-based learning server DS can generate training models corresponding to each of the two-stage cooling mode, the tropical sound night mode, and the tropical sound night sleep mode in the calm mode as the operating mode of the room air conditioner.

[161] If each training model according to the training procedure 404 considering the air conditioner A type 401 and the air conditioner A mode 402 is formed, the data-based learning server DS may use the training model to obtain the recommended temperature. In this case, the recommended temperature can be obtained considering the setting range 405 for each operating mode. For example, when the recommended temperature obtained by the data learning server DS is outside the setting range 405, the temperature in the setting range 405 closest to the recommended temperature may be determined as the final recommended temperature.

[162] When describing an example of the training model generation procedure in the intelligent comfort mode 411 with reference to FIG. 4, room temperature (or current temperature) and desired temperature (or set temperature) can be used. In this case, the room temperature may be the room temperature measured at the time of changing the desired temperature. In addition, data imported over a specific period of time can be used as training data. A particular time period, for example, could be data imported in a particular year, a particular month, or a particular season. The specific data may be data imported based on the air conditioner temperature setting history information of unspecified users who use products identical or similar to air conditioner A, as well as air conditioner user A. At this time, unspecified users may be restricted, for example, to a user in the same or similar area or identical or similar environment with air conditioner A.

[163] In the intelligent comfort mode 411, the data learning server DS may use the current temperature (or room temperature), the outside temperature, the outside humidity, and the desired temperature as the learning data.

[164] In addition, the data-based learning server DS may use the external environment information based on the local information of the air conditioner A as training data. On the other hand, when the data learning server DS may not confirm the local information of the air conditioner A, the data learning server may generate, train, and replace with a new training model using the current temperature and the desired temperature as training data.

[165] The data-based learning server DS can obtain the recommended temperature to be set in the air conditioner by using the generated, trained and replaced with new learning models.

[166] In this case, if the received recommended temperature is outside the setting range of 22-26°C, the learning server DS may determine the final recommended temperature based on the setting range based on the data.

[167] For example, if the recommended temperature obtained using the training model is less than 22°C, the data-based learning server DS may determine the recommended temperature to be 22°C. In addition, if the recommended temperature obtained using the training model is greater than or equal to 26°C, the data-based learning server DS may determine the recommended temperature to be 26°C.

[168] According to various embodiments, when generating the training model, the data-based learning server DS may further assign a weight to the training data newly imported to generate the training model.

[169] FIG. 5 is a diagram showing an example of applying a weighting factor to training data according to an embodiment of the disclosure.

[170] Referring to FIG. 5, the data-based learning server DS may assign weights to the training data imported for 1 day, 2 days, and 3 days, respectively, as 501, 502, and 503 in FIG. 5.

[171] For example, at 501 of FIG. 5, the data-based learning server DS may assign a weight of 0.8 to all data (for example, data imported from unspecified users) over the past year, and assign a weight of 0.2 to the user's personal data (desired temperature and current temperature of the user and etc.) of air conditioner A, which are imported on the first day. Similarly, at 502 of FIG. 5, the data-based learning server DS may assign a weight of 0.8 to all data in the past year and the personal data of the user of the air conditioner A that are imported on the first day, and a weight of 0.2 to the personal data of the user of the air conditioner A that are imported on the first day. second day. In addition, at 503 of FIG. 5, the data-based learning server DS can assign a weight of 0.8 to all of the last year's data and the personal data of the user of the air conditioner A that are imported on the first day and the second day, and a weight of 0.2 to the user's personal data that are imported. on the third day.

[172] On the other hand, the above weight value is only an example, and the data-based learning server DS may be preset according to various values by a manufacturer, a manager, an operating system, an application provider, or the like. for the data-based learning server DS. For example, in FIG. 5, instead of a weighting factor of 0.8 and a weighting factor of 0.2, a weighting factor of 0.9 and a weighting factor of 0.1 may be used. As another example, in FIG. 5, instead of a weighting factor of 0.8 and a weighting factor of 0.2, a weighting factor of 0.7 and a weighting factor of 0.3 may be used.

[173] On the other hand, the above weighting factor may be a variable type that changes depending on the situation, rather than a predetermined fixed type.

[174] In this case, the weighting factor may be changed manually by an administrator of the data-based learning server DS, an air conditioner user, or the like. or it can automatically change depending on the specific condition. For example, as the total volume of imported training data increases, the weight of the most recently imported personal data may also increase accordingly.

[175] FIG. 6 is a diagram showing the structure of a cloud server according to an embodiment of the disclosure.

[176] Cloud server C may include a batch processing unit 601, a content provider (CP) data collection unit 602, a CSV format creation unit 603, a model creation unit 604, and a temperature providing server API 605 for training. The components 601 to 604 of the cloud server C described above use and process the data stored in the storage unit (or database) of the cloud server C to generate a recommended temperature suitable for the air conditioner A.

[177] First, the cloud server C may store, in the device state storage unit 651, the device state data including the state information of the air conditioner A obtained depending on the generation of the state change event of the air conditioner A. The device state storage unit 651 may correspond to , for example, the device status data DB BS2 of FIG. 1A and 1B. The cloud server C may acquire the state information stored in the device state storage unit 651 every predetermined period (for example, every day), and stores the raw data generated depending on a certain criterion (for example, by date) in the object storage unit 652 .

[178] The C cloud server batch processing unit 601 may receive and filter the row data in the object storage unit 652 and store the filtered data in the distributed environment data DB 653 (eg, Not Only SQL DB, NoSQL DB). The filtered data, for example, may be data including state data of the device of air conditioner A or state information extracted from metadata.

[179] In addition, the CP data acquisition unit 602 may store weather data including weather information imported from the external content server CP in the object storage unit 654 .

[180] The C cloud server CSV format generating unit 603 refines the data obtained from the distributed object environment data DB 653 and the object storage unit 654 to generate specific format data (eg, CSV format) suitable for generating a training model and store generated data in the object storage unit 655 .

[181] Model building unit 604 may receive data of a specific format from object storage unit 655, generate a training model using the data, and store the generated training model in object storage unit 656 .

[182] The cloud server C may temporarily store the training model stored in the object storage unit 655 in the cache 657, which is a high-speed storage device, when the use of the training model is required.

[183] In a situation in which the use of a training model is required, the recommended temperature providing API 605 of the cloud server C can obtain the recommended temperature of the air conditioner A by using the training model stored in the cache 657.

[184] The cloud server C may transmit the recommended temperature obtained through the received recommended temperature providing API 605 to the mobile applications of the air conditioner A and the user terminal U.

[185] Meanwhile, in FIG. 6, for ease of explanation, object storage units 652, 654, 655, and 656 are designated by various reference numbers, but object storage units 652, 654, 655, and 656 may denote an identical object storage unit, or may denote two or more distributed object storage units. .

[186] FIG. 7A and 7B are diagrams showing a training model generation procedure according to an embodiment of the disclosure.

[187] The training model may be generated using an artificial intelligence algorithm. For example, the training model may be generated using a decision tree algorithm, a support vector machine algorithm, a linear discriminative analysis algorithm, a genetic algorithm, or a neural network algorithm that models neurons in a human neural network. The neural network algorithm may include a plurality of network nodes having weights. A plurality of network nodes can interconnect connections in such a way that neurons model the synaptic activity of transmitting and receiving signals across synapses. In addition, the training model can be formed using a deep learning algorithm developed in a neural network algorithm. In a deep learning algorithm, multiple network nodes can transmit and receive data depending on the convolutional interconnection of connections when located at different depths (or layers). The training model may include models such as deep neural network (DNN), recurrent neural network (RNN), and bidirectional recurrent deep neural network (BRDNN) may be provided, but are not limited to the above example.

[188] For ease of description, the disclosure describes a method for providing a recommended temperature using linear regression as the algorithm used to generate the training model.

[189] The data-based learning server DS may extract a training model, for example, the following Equation 1, according to a linear regression algorithm.

[190] y=a0+a1×1+a2×2+a3×3... equation 1

[191] In the above equation 1, y is a variable associated with the set temperature set in the air conditioner A, and a0, a1, a2 and a3 are constant values. In addition, x1 is a variable related to the current temperature, x2 is a variable related to the outdoor temperature, and x3 is a variable related to the outdoor humidity.

[192] To simplify understanding, the training model in the case where the number of training variables (or training items) in the above equation 1 is two is expressed by the following equation 2.

[193] y=a0+a1×1... equation 2

[194] In this case, the table of FIG. 7A shows the set temperature 712 (eg, user setting temperature) versus the current temperature 711 (or ambient temperature, room temperature) of the air conditioner.

[195] Based on the linear regression algorithm, the data-based learning server DS may derive a training model, which is a calculation expression that expresses the relationship of the set temperature 712 as a function of the current temperature 711.

[196] This is shown in the graph shown in FIG. 7B.

[197] Referring to FIG. 7B, current temperature 711 and set temperature 712 in FIG. 7A may correspond to the X label on the graph when they are illustrated on the X and Y axes.

[198] In this case, the linear regression line 721 using the linear regression algorithm can be obtained such that the error sum of the marker set X is small. Thus, in the above Equation 2, the constant values a0 and a1 having the smallest difference between the set temperature 712 of air conditioner A and the predicted temperature can be calculated.

[199] An example of a linear regression model that reflects the calculated constant value is as follows.

[200] y=29.91840623+(–0.3717125) x1... equation 3

[201] Accordingly, the data-based learning model DS can provide the air conditioner A with the recommended temperature according to the air conditioner A recommended temperature request command based on the following Equation 3.

[202] For example, when the current room temperature near air conditioner A is 26°C, the recommended temperature provided by using the training model of Equation 3 above may be 19°C.

[203] According to various embodiments, the training model may be continuously replaced with a new one (or updated).

[204] To this end, the data-based learning server DS may further include a new model replacement unit (not shown). The new model replacement unit may determine whether or not the training model has been replaced with a new one by analyzing the relevance between the base training data used in the training model that is constructed in advance and the newly input training data. At this time, the relevance may be determined based on the region and time at which the training data is generated, the time, the model of the air conditioner that provides the training data, and the like.

[205] For example, the replacement unit can continuously replace with a new training model already constructed by using the user's temperature setting history to set the temperature of the air conditioner A, the user's change history for the recommended temperature, and the like. as training data.

[206] According to various embodiments, the training model may be stored in the storage unit of air conditioner A rather than in a separate server. In this case, the training model constructed on the data-based learning server DS may be transmitted to the conditioner A periodically or when an event is generated.

[207] When the training model is provided in the air conditioner A, the air conditioner A can obtain the recommended temperature using the stored training model. For example, air conditioner A can obtain the recommended temperature by inputting the read current temperature into the training model. In this case, air conditioner A can obtain the recommended temperature using the read current temperature without user intervention, and can automatically set the temperature of air conditioner A according to the recommended temperature.

[208] FIG. 8 is a block diagram showing a configuration of an air conditioner according to an embodiment of the disclosure.

[209] Referring to FIG. 8, air conditioner A may include a temperature sensor 810, a fan 820, a communication unit 830, a storage unit 840, and a processor 850. In various embodiments, air conditioner A may not include at least one of the components described above, or may optionally include other components.

[210] Temperature sensor 810 can sense the temperature of a room near air conditioner A.

[211] The fan 820 may exhaust cooling air to the outside through an opening/closing portion (not shown). Alternatively, in calm mode, fan 820 may exhaust cooling air to the outside through a plurality of micro-cavities (not shown) at a predetermined flow rate or less. At this time, the predetermined flow rate may be 0.25 m/s or less, preferably 0.15 m/s or less.

[212] The communication unit 830 may communicate with an external device. At this time, the external device may include at least one of the cloud server C, the data-based learning server DS, and the user terminal U. Communication of the communication unit 830 with the external device may include communicating with the external device via the third device and etc. For example, communication unit 830 may receive a remote control signal to control air conditioner A from user terminal U.

[213] The communication unit 830 may communicate with an external device via wired communication or wireless communication. For example, the communication unit 830 can communicate with the control terminal device via cellular communication, near field communication and the Internet, as well as a cable connection port, and communicate according to standards such as universal serial bus (USB) communication. , Wi-Fi, Bluetooth technology, ZigBee, Infrared Data Association (IrDA), RF such as UHF and VHF, and ultra-wideband (UWB) communications.

[214] The storage unit 840 stores various software and programs for performing the function of air conditioner A. In particular, the storage unit 840 can store a temperature control algorithm according to a plurality of operating modes. The temperature control algorithm may include changing the set temperature, wind speed intensity, wind speed direction, and the like. depending on a predetermined period for each operating mode. Further, according to the disclosure, the storage unit 840 may store the trained training model based on the set temperature and the current temperature.

[215] The processor 850 may read the program and the like stored in the data storage unit 840.

In particular, in order to perform the function of the air conditioner A, the processor 850 may read programs including the sequence of instructions to be read, and perform air conditioning according to a predetermined temperature.

[216] The processor 850 may read the pressure and/or temperature of the refrigerant in the indoor heat exchanger (not shown) to read whether or not air conditioning is performed normally. For example, the processor 850 can read whether the indoor heat exchanger tube is damaged or frosted or not, and whether or not the water formed by condensing the vapor in the air is properly removed or not.

[217] The processor 850 may control the speed of the fan 820. In particular, the processor 850 may control the current temperature measured by the temperature sensor 810 and the speed at which the fan 820 rotates depending on the set temperature. In particular, the processor 850 may control the speed at which the fan 820 rotates in response to the difference between the current temperature and the set temperature. For example, if the difference between the current temperature and the set temperature is large, the rotation speed of the fan 820 is controlled to be high so as to quickly reach the set temperature, and if the difference between the room temperature and the set temperature is small or the room temperature reaches the set temperature , the room temperature also drops excessively, the rotation speed of the fan 820 may be slow so that the compressor of the outdoor unit does not turn off. For example, processor 850 may control the speed of fan 820 between 500 rpm and 900 rpm.

[218] The processor 850 may control the communication unit 830 to transmit the current temperature and the set temperature read by the temperature sensor 810 to an external device.

[219] In addition, the processor 850 can control the communication unit 830 to receive the recommended temperature received from the external device, and control the recommended temperature received through the communication unit 830 so that it is set in the air conditioner as the set temperature.

[220] In addition, the processor 850 may control the communication unit 830 to transmit the current temperature read by the temperature sensor 810 to the external device, and may receive the recommended temperature depending on the transmission of the current temperature from the external device, and set the received recommended temperature. in the air conditioner. In this case, the recommended temperature may be the result of applying the current temperature read by the temperature sensor 810 to the trained training model using the set temperature set and the current temperature set in air conditioner A. In this case, the external device may include at least one of the cloud server C, the training model server DS, and a third device operatively connected to the cloud server C or the training model server DS.

[221] According to various embodiments, a network system can be provided that includes conditioner A and a training model server DS generating a training model using training data received from conditioner A.

[222] In this case, the network system air conditioner A may include a temperature sensor 810 that reads the current temperature, a fan 820 that exhausts cooling air to the outside, and a communication unit 830 capable of communicating with an external device. The air conditioner A may include a processor 850 that controls the communication unit 830 to transmit the set temperature set in the air conditioner A and the current temperature read by the temperature sensor 810 to an external device.

[223] In this case, the external device may include at least one of the cloud server C, the training model server DS, and a third device operatively connected to the cloud server C or the training model server DS.

[224] In addition, the network system training model server DS may include a training data acquisition unit (for example, the training data acquisition unit 203a in FIG. 2A) that obtains the current temperature and the set temperature transmitted from the air conditioner A, the training model-based learning unit (eg, model-based learning unit 203b of FIG. 2A) that generates a training model using a set temperature and current temperature, and a data storage unit (eg, data storage unit 202 of FIG. 2A) that stores the trained training model. model to provide the recommended temperature of air conditioner A as a result of the formation of the training model.

[225] According to various embodiments, a network system can be provided that includes air conditioner A and a training model server DS providing a recommended temperature using recognition data received from air conditioner A.

[226] In this case, the network system air conditioner A includes a temperature sensor 810 sensing the current temperature, a fan 820 exhausting cooling air to the outside, a communication unit 830 capable of communicating with an external device, and a processor 850 controlling the communication unit 830 so as to transmit the current temperature read by the temperature sensor 810 to an external device.

[227] In this case, the external device may include at least one of the cloud server C, the training model server DS, and a third device operatively connected to the cloud server C or the training model server DS.

[228] In addition, the training model server DS may include a data storage unit (for example, data storage unit 202 in FIG. 2B) storing trained training models to provide the recommended temperature of air conditioner A, a recognition data acquisition unit (for example, a 203c of obtaining the recognition data of Fig. 2B) obtaining the current temperature of the air conditioner A, and a model application unit (for example, the model application unit 203d of Fig. 2B) obtaining the recommended temperature of the air conditioner A by inputting the current temperature as a training model, and a communication unit (eg, communication unit 201 of FIG. 2B) transmitting the received recommended temperature to an external device. The external device may include an air conditioner A or a third device operatively connected to the air conditioner A. In addition, the external device may include a user terminal U or a third device operatively connected to the user terminal U to transmit the recommended temperature.

[229] FIG. 9 is a block diagram showing a configuration of a user terminal according to an embodiment of the disclosure.

[230] Referring to FIG. 9, the user terminal U may include a display 910, a communication unit 920, an input 930, a data storage unit 940, and a processor 950.

[231] The display 910 may visually provide information to the user of the user terminal U. For example, the display 910 may display a screen including an AI operation UI controlled by the processor 950.

[232] The communication unit 920 can establish a channel for a wired or wireless communication unit between the user terminal U and an external device, and maintain communication performance through the established communication channel. The external device may include at least one of the cloud server C, the training model server DS, and a third device operatively connected to the cloud server C or the training model server DS, for example.

[233] The communication unit 920 can communicate with an external device via near-field communication networks (for example, Bluetooth technology, Wi-Fi Direct standard or Infrared Data Association (IrDA) standard, etc.) or remote networks communication (for example, a cellular network, an Internet network or a computer network (for example, a LAN or WAN), etc.) using wireless communication modules (for example, a cellular communication module, a local wireless communication module, and a global navigation satellite communication module). system (GNSS)) or a wired communication module (for example, a local area network (LAN) communication module or a power line communication module). The several kinds of communication modules described above may be implemented as a single chip or may be implemented as a separate chip.

[234] Input 930 may receive commands or data to be used for components (eg, processor 950) of user terminal U from outside (eg, from the user) of user terminal U. Input 930 may include, for example, a button, a microphone, touch panel, etc. The input 930 may transmit a user input signal generated depending on the user input for controlling the user terminal U to the processor 950.

[235] The data storage unit 940 may store various data used by at least one component (eg, processor 950) of the user terminal U, such as software (eg, a program), and may store input data or output data for an instruction, associated with them. The data storage unit 940 may include volatile and/or non-volatile storage.

[236] The program is software stored in the storage unit 940 and may include, for example, an operating system, middleware, or an application.

[237] Processor 950 can direct, for example, software (eg, program) stored in storage unit 940 to control at least other components (eg, hardware or software components) of user terminal U coupled to processor 950 , and perform various data processing and operations. Processor 950 may load an instruction or data received from other components (eg, communication unit 920) to non-volatile storage and process the downloaded instruction or data and may store the resulting data in non-volatile storage. According to an embodiment, processor 950 may include main processors (eg, a CPU or application processor) and sub-processors (eg, a graphics processor, an image signal processor, a sensor hub processor, or a communications processor) that operate independently of the main processor and additionally or alternatively use a lower power level than the main processor, or are designed specifically for the specified functions. The subprocessor may operate separately from the main processor or may operate inline.

[238] According to various embodiments, if a user input signal dependent on the user input selecting the AI operation UI included in the screen provided by the display 910 is received via the input 930, the processor 950 may control the communication unit 920 to transmit the request signal of the AI operation corresponding to the AI operation UI to the air conditioner A. If the recommended temperature set in the air conditioner depending on the AI operation request signal is obtained through the communication unit 920, the processor 950 may control the display 910 to display the obtained recommended temperature . At this time, the recommended temperature may be obtained as a result obtained by allowing air conditioner A to apply the current temperature of air conditioner A to the training model. In this case, the processor 950 may control the display 910 such that the user displays the set temperature that was set in the air conditioner in the past as equal to the current temperature along with the recommended temperature.

[239] According to various embodiments, a network system can be provided including an air conditioner A and a user terminal U controlling the air conditioner A.

[240] In this case, if the user input signal dependent on the user input selecting the AI operation UI included in the screen provided by the display 910 of the user terminal U is received via the input 930, the processor 950 may control the communication unit 920 to transmit the AI operation request signal corresponding to the AI operation UI to the air conditioner A.

[241] If the air conditioner A receives the AI operation request via the air conditioner A communication unit 830, the air conditioner A processor 850 may control the communication unit 830 to transmit the current temperature of the air conditioner A to an external device. The processor 850 of air conditioner A may control the communication unit 830 to receive the recommended temperature depending on the transmission of the current temperature from the external device. The processor 850 may set the recommended temperature received via the communication unit 830 in the air conditioner A. In this case, the recommended temperature may be the result obtained by applying the current temperature to the trained training model based on the set of set temperatures preset in the air conditioner A, and set of current temperatures. In this case, the external device may include at least one of the cloud server C, the training model server DS, and a third device operatively connected to the cloud server C or the training model server DS.

[242] FIG. 10A and 10B are diagrams showing a user terminal screen displaying a recommended temperature, according to an embodiment of the disclosure.

[243] Referring to FIG. 10A, the user terminal U may display a screen 1010 for controlling the air conditioner by executing an application capable of controlling the air conditioner A.

[244] The screen 1010 for controlling the air conditioner may include UI 1011 turning on/off air conditioner A, UI 1012 selecting the operating mode of air conditioner A, current temperature information 1013, information 1014 regarding whether the artificial intelligence mode is operating or not, setting UI 1015 wind damper, wind force setting UI 1016, no-wind function UI 1017, UI 1018 regarding whether or not air cleaning operation is performed, artificial intelligence setting UI 1019, backup setting UI 1020, and the like.

[245] In this case, when the air conditioner control screen 1010 is outside the viewport range of the UI display of the user terminal, the user can display the air conditioner control screen 1110 that is outside the viewport range within the viewport range by a drag gesture.

[246] Referring to FIG. 10A and 10B, in this situation, if the user input for selecting the AI setting UI 1019 is received, the user terminal U can display the AI control screen 1020 in the operation mode (e.g., intelligent comfort mode) of the air conditioner A, as shown in FIG. 10b. The AI control screen 1020 may include AI operation UI 1021 for AI operation of air conditioner A and AI operation information 1022 indicating AI operation of air conditioner A.

[247] If the user input for selecting the AI operation UI 1021 is received, the user terminal U can obtain the recommended temperature set in the air conditioner based on the user input. For example, the user terminal U may receive the recommended temperature through a third device (eg, an access point (AP)) operatively connected to the cloud server C.

[248] Then, the user terminal U may display the recommended temperature 1023 on the AI control screen 1020. At this time, the recommended temperature 1023 can be obtained as a result obtained by allowing the air conditioner A to apply the current temperature of the air conditioner A to the training model server DS, based on the user input selecting the AI operation UI 1021.

[249] The user terminal U can display not only the recommended temperature 1023 on the AI control screen 1020, but also the set temperature 1024 that the user of the air conditioner A directly sets in the air conditioner in the past. In this case, the recommended temperature 1023 and the set temperature 1034 can be displayed together on the graph so that they can be compared with each other.

[250] FIG. 11 is a flowchart showing a method for generating a training model for a learning server based on data according to an embodiment of the disclosure.

[251] Referring to FIG. 11, in step 1101, the data-learning server DS may obtain the set temperature set in the air conditioner A and the current temperature of the air conditioner A at the time of temperature setting. Further, the data-based learning server DS may further acquire the outside environment information of the air conditioner A.

[252] At this time, the data-based learning server DS can obtain the set temperature and the current temperature from the bridge server BS operatively connected to the air conditioner A, and obtain external environment information from the intelligent home service server SS operatively connected to the external server ( CP) providing content.

[253] Additionally, the data-based learning server DS may further obtain time information at the time of setting the temperature in the air conditioner A.

[254] In step 1103, the data-based learning server DS may generate or replace a new learning model using the obtained target temperature and the current temperature.

[255] When the data-based learning server DS further obtains the external environment information, the data-based learning server DS may generate or replace with a new training model using the acquired target temperature, the current temperature, and the external environment information.

[256] In addition, when the data-based learning server DS further obtains time information at the time of temperature setting, the data-based learning server DS may generate or replace with a new training model using the obtained target temperature, current temperature, and time information.

[257] In step 1105, the data-based learning server DS may store the trained learning model to provide the recommended temperature to be set in the air conditioner by generating and replacing with a new learning model.

[258] Meanwhile, the data-based learning server DS may generate or replace a plurality of training models for each operating mode of the air conditioner A. In this case, the data-based learning server DS may store a plurality of training models.

[259] FIG. 12 is a flowchart showing a method of using a training model for a data-based learning server according to an embodiment of the disclosure.

[260] Referring to FIG. 12, in step 1201, the data-based learning server DS may store the trained learning model to provide the recommended temperature to be set in air conditioner A.

[261] In a situation in which the trained training model is stored, in step 1203, the data-based learning server DS may obtain the current temperature of the air conditioner A. In this case, the data-based learning server DS may further obtain the outside environment information of the air conditioner A.

[262] In step 1205, the data-based learning server DS may input the obtained current temperature into the trained learning model to obtain a recommended temperature to be set in air conditioner A.

[263] In addition, when the data-based learning server DS further obtains the external environment information, the data-based learning server DS may input the obtained recommended temperature and the external environment information into the training model to obtain the recommended temperature to be set in the air conditioner A.

[264] Meanwhile, the data-based learning server DS may store a plurality of training models for each operating mode of air conditioner A. In this case, the data-based learning server DS may input the acquired current temperature into the training model corresponding to the current operating mode of air conditioner A, and input the obtained current temperature into the training model corresponding to the current operating mode of air conditioner A to obtain the recommended temperature to be set in air conditioner A.

[265] In step 1207, the data-based learning server DS may transmit the obtained recommended temperature to an external device. The external device, for example, may be air conditioner A or a third device operatively connected to air conditioner A to transmit the recommended temperature. In addition, the external device may be a user terminal U or a third device operatively connected to the user terminal U to transmit the recommended temperature.

[266] FIG. 13 is a flowchart showing a method for providing the recommended temperature of air conditioner A according to an embodiment of the disclosure.

[267] Referring to FIG. 13, in step 1301, air conditioner A may read the current temperature of air conditioner A.

[268] Next, in step 1303, the air conditioner A may transmit the read current temperature to the external device. For example, the air conditioner A may transmit the sensed current temperature to at least one of the cloud server C, the training model server DS, and a third device operatively communicating with the cloud server C or the training model server DS.

[269] In step 1305, the air conditioner A may receive the recommended temperature, which is the result of applying the current temperature to the training model, from the external device depending on the current temperature being transmitted. In this case, the recommended temperature may be a result obtained by applying the current temperature to the trained training model based on the set temperature set in advance in the air conditioner A and the current temperature set.

[270] In step 1307, air conditioner A may set the air conditioner to receive recommended temperature.

[271] FIG. 14 is a flowchart showing a method for controlling an air conditioner of a user terminal according to an embodiment of the disclosure.

[272] Referring to FIG. 14, in step 1401, the user terminal U may receive a user input signal depending on the user input selecting the AI operation UI.

[273] In step 1403, the user terminal U may transmit an AI operation request signal corresponding to the AI operation UI to the air conditioner A in response to the user input signal.

[274] In step 1405, the user terminal U may obtain the recommended temperature set in the air conditioner by applying the current temperature of the air conditioner A to the training model, depending on the AI operation request signal.

[275] In step 1407, the user terminal U may display the received recommended temperature on the screen. In this case, the user terminal U may display the set temperature that the user presets in the air conditioner together with the recommended temperature as equal to the current temperature.

[276] FIG. 15 is a flowchart of a method in a network system including a user terminal and an air conditioner according to an embodiment of the disclosure.

[277] Referring to FIG. 15, in step 1501, the user terminal U may receive a user input signal depending on the user input selecting the AI operation UI.

[278] In step 1503, the user terminal U may transmit an AI operation request signal corresponding to the AI operation UI to the air conditioner A.

[279] In step 1505, air conditioner A may read the current temperature of air conditioner A.

[280] Next, in step 1507, air conditioner A may transmit the read current temperature to the external device 1500. The external device may include at least one of cloud server C, training model server DS, and a third device operatively connected to cloud server C or the training model server DS.

[281] At step 1509, the air conditioner A may receive the recommended temperature, which is the result of applying the current temperature to the training model, from the external device 1500 depending on the transmission of the current temperature. In this case, the training model may be a training model trained using the set temperature set in advance in air conditioner A and the current temperature set.

[282] In step 1511, air conditioner A may set the air conditioner to receive recommended temperature.

[283] The disclosed embodiments may be implemented as a software program that includes instructions stored on a computer-readable storage medium.

[284] A computer is equipment that calls stored instructions from a storage medium, and may operate according to the disclosed embodiment depending on the instructions being called, and may include a data-based learning server according to the disclosed embodiments, or an external server operatively connected to data-based learning server. Alternatively, the computer may include an air conditioner or an external server operatively connected to the air conditioner according to the disclosed embodiments.

[285] A computer-readable storage medium may be provided as a non-transitory storage medium. "Non-temporal" means that the storage medium does not include a signal and current and is tangible, and "non-temporal" does not distinguish whether the data is stored semi-permanently or temporarily on the storage medium. As an example, non-temporary storage media can be temporarily stored temporary storage media such as registers, cache and buffer, as well as non-temporary readable media such as CD, DVD, hard disk, Blu-ray disc, USB, internal storage device, memory card, ROM and RAM.

[286] In addition, the method according to the disclosed embodiments may be provided as a computer program product.

[287] A computer program product may include a software program, a computer-readable storage medium on which the software program is stored, or a product traded between a seller and a buyer.

[288] For example, a computer program product may include a product (e.g., a downloadable application) in the form of a software program electronically distributed through a data-based learning server, through an air conditioner manufacturer, or through an electronic marketplace (e.g., Google Play Store, AppStore ). For electronic distribution, at least a portion of the firmware programs may be stored on a storage medium or may be temporarily generated. In this case, the storage medium may be a manufacturer or an electronic marketplace server or a relay server storage medium.

[289] Although embodiments of the disclosure have been illustrated and described, the disclosure is not limited to the above specific embodiment, but may be modified in various ways by those skilled in the art to which the disclosure relates without departing from the spirit and scope of the disclosure as claimed in the claims. inventions. In addition, it should be understood that such modifications fall within the scope of the disclosure.

[290] Although the disclosure has been shown and described with reference to various embodiments thereof, those skilled in the art should understand that various changes in form and content may be made without departing from the spirit and scope of the disclosure as defined by the appended claims and its equivalents.

Claims (182)

1. Data-based learning server, containing: communication block; data storage unit; and at least one processor configured to: receive the set temperature set in the air conditioner and the current ambient temperature in the air conditioner room during the setting of the set temperature through the communication unit, apply the set temperature and the current room ambient temperature to the training model, and receive the recommended temperature set in the air conditioner, wherein the training model performs training to provide the recommended temperature based on the set temperature and the current room ambient temperature. 2. The data-based learning server of claim 1, wherein at least one processor is further configured to: receive information from the external environment, and generate or update the training model using the set temperature, the current room ambient temperature, and the outside environment information. 3. The learning server based on the data of claim 2, wherein the environmental information comprises at least one of an outdoor temperature and an outdoor humidity at the time of setting the set temperature. 4. The data-based learning server of claim 2, wherein at least one processor is further configured to: receive the set temperature and the current room ambient temperature from the bridge server communicatively connected to the air conditioner via the communication unit, and receive external environment information from the intelligent home service server, which is communicatively connected to the external content provision server via the communication unit. 5. The data-based learning server of claim 1, wherein at least one processor is further configured to: receive time information at the time of setting the set temperature, and generate or update a training model using the setpoint temperature, current room temperature, and time information. 6. Learning server based on the data according to claim 1, wherein at least one processor is further configured to generate or update a plurality of training models for each operating mode of the air conditioner, and wherein the data storage unit is further configured to store a plurality of training models. 7. Data-based learning server, comprising: a data storage unit configured to store the trained training model to provide a recommended temperature to be set to the air conditioner; at least one processor configured to: get the current ambient temperature in the air conditioner room, apply the set temperature and the current room ambient temperature to the trained training model, and receive the recommended temperature set in the air conditioner; and a communication unit configured to transmit the recommended temperature to an external device, wherein the trained training model performs training to provide a recommended temperature based on the set temperature and the current room ambient temperature. 8. The data-based learning server of claim 7, wherein at least one processor is further configured to: receive information from the external environment, and input the current room ambient temperature and the outside environment information into the trained training model to obtain the recommended temperature to be set in the air conditioner. 9. The data-based learning server of claim 7, wherein when the data storage unit stores a plurality of training models for each operating mode of the air conditioner, at least one processor is further configured to input the current room ambient temperature into the trained training model, corresponding to the current operating mode of the air conditioner to obtain the recommended temperature of the air conditioner. 10. Conditioner, containing: a fan configured to discharge cooling air to the outside; a temperature sensor configured to read the current ambient temperature in the room near the air conditioner; a communication unit configured to communicate with an external device; and at least one processor configured to: control the communication unit so as to transmit the set temperature set in the air conditioner and the current ambient temperature in the air conditioner room during the setting of the set temperature through the communication unit, control the communication unit so as to receive the recommended temperature, which is the result obtained by applying the set temperature in the training model, and set the accepted recommended temperature in the air conditioner, wherein the training model performs training to provide the recommended temperature based on the set temperature and the current room ambient temperature. 11. The user terminal that controls the air conditioner, and the user terminal contains: a display configured to display the screen; a communication unit configured to communicate with an external device; an input receiving device configured to receive user input; and at least one processor configured to: control the communication unit to transmit a control command requesting the activation of the artificial intelligence mode of the air conditioner corresponding to the user interface (UI) of the artificial intelligence operation to the air conditioner in response to the user input signal depending on the user input selecting the UI of the artificial intelligence operation included in the screen, which is received through the input receiver, and control the display to show the recommended temperature in response to the recommended temperature set in the air conditioner, which is the result obtained by: receiving the set temperature set in the air conditioner and the current ambient temperature in the air conditioner room during the setting of the set temperature through the communication unit, applying the set temperature and the current ambient temperature in the air conditioner room in the training model, and obtaining the recommended temperature set in the air conditioner, wherein the training model performs training to provide the recommended temperature based on the set temperature and the current room ambient temperature. 12. The user terminal of claim 11, wherein the processor is further configured to control the display to display the set temperature preset in the air conditioner as equal to the current room ambient temperature along with the recommended temperature. 13. Network system, containing: air conditioning; and a training model server configured to generate a training model using training data received from the air conditioner, while the conditioner contains: a fan configured to discharge cooling air to the outside, a temperature sensor configured to read the current ambient temperature in the room near the air conditioner, a communication unit configured to communicate with an external device, and at least one air conditioner processor configured to control the communication unit to transmit the set temperature set in the air conditioner and the current room ambient temperature read by the temperature sensor to the external device, and while the training model server contains: at least one server processor configured to: receive the set temperature set in the air conditioner and the current ambient temperature in the air conditioner room during the setting of the set temperature, apply the set temperature and the current room ambient temperature to the training model, and receive the recommended temperature set in the air conditioner, and wherein the training model performs training to provide the recommended temperature based on the set temperature and the current room ambient temperature. 14. Network system, containing: air conditioning; and a training model server configured to provide the recommended temperature using recognition data received from the air conditioner, while the conditioner contains: a fan configured to discharge cooling air to the outside, a temperature sensor configured to read the current ambient temperature in the air conditioner room, an air conditioner communication unit configured to communicate with an external device, and at least one air conditioner processor configured to control the air conditioner communication unit to transmit the current room ambient temperature read by the temperature sensor to an external device, and while the server of training models contains: a data storage unit configured to store the trained training model to provide the recommended air conditioner temperature, at least one server processor configured to: get the set temperature set in the air conditioner and the current ambient temperature in the room, apply the set temperature and the current room ambient temperature to the trained training model, and receive the recommended temperature set in the air conditioner, and a server communication unit configured to transmit the recommended temperature to an external device, and wherein the trained training model performs training to provide a recommended temperature based on the set temperature and the current room ambient temperature. 15. Network system, containing: air conditioning; and a user terminal configured to control the air conditioner, while the user terminal contains: a display configured to display a screen, a terminal communication unit configured to communicate with an external device, an input receiver configured to receive user input, and at least one terminal processor configured to control the communication unit of the terminal so as to transmit a control command requesting the activation of the artificial intelligence mode of the air conditioner corresponding to the user interface (UI) of the artificial intelligence operation to the air conditioner in response to a user input signal depending on a user input selecting an artificial intelligence operation UI included in a screen that is received through an input receiver, while the conditioner contains: a fan configured to discharge cooling air to the outside, a temperature sensor configured to read the current ambient temperature in the room near the air conditioner, an air conditioner communication unit configured to communicate with an external device, and at least one air conditioner processor configured to: control the communication unit of the air conditioner to transmit the current ambient temperature in the room to an external device, receive the recommended temperature depending on the transmission of the current room ambient temperature from the external device in response to the control command requesting the activation of the artificial intelligence mode of the air conditioner received through the air conditioner communication unit, and set the accepted recommended temperature in the air conditioner, whereby the recommended temperature is the result obtained by: obtaining the set temperature set in the air conditioner and the current ambient temperature in the air conditioner room at the time of setting the set temperature, applying the set temperature and the current room ambient temperature in the trained training model based on the set of set temperatures previously set in the air conditioner and the set of current room ambient temperatures, and obtain the recommended temperature set in the air conditioner, and wherein the trained training model performs training to provide the recommended temperature based on the set temperature and the current room ambient temperature. 16. A method for generating a training model for a training server based on data, the method comprising the steps of: obtaining the set temperature set in the air conditioner and the current ambient temperature in the air conditioner room at the time of setting the set temperature, apply the set temperature and the current ambient room temperature in the training model, and receive the recommended temperature set in the air conditioner, and save the training model to provide the recommended temperature set in the air conditioner as a result of applying the set temperature and the current ambient temperature in the room in the training model, wherein the training model performs training to provide the recommended temperature based on the set temperature and the current room ambient temperature. 17. The method of claim 16, further comprising the step of: receive information about the external environment of the air conditioner, wherein generating or updating the training model comprises generating or updating the training model using the set temperature, the current ambient temperature in the room, and the external environment information. 18. The method of claim 17, wherein the ambient information comprises at least one of an outdoor temperature and an outdoor humidity at the time of setting the set temperature. 19. The method according to p. 17, wherein obtaining the set temperature and the current room ambient temperature comprises obtaining the set temperature and the current room ambient temperature from a bridge server communicatively connected to the air conditioner, and wherein obtaining the external environment information comprises the step of obtaining the external environment information from the intelligent home service providing server, which is connected with the possibility of communication with the external content providing server. 20. The method of claim 16, further comprising the step of: receive time information during the setting of the set temperature, wherein generating or updating the training model comprises generating or updating the training model using the set temperature, the current room ambient temperature, and the time information. 21. The method according to p. 16, wherein generating or updating the training model comprises generating or updating a plurality of training models for each operating mode of the air conditioner, and wherein saving the training model comprises the step of storing a plurality of training models. 22. A method for using a training model for a data-based learning server, said method comprising the steps of: storing the trained training model to provide the recommended temperature to be set to the air conditioner, get the set temperature set in the air conditioner and the current ambient temperature in the air conditioner room, apply the set temperature and the current room ambient temperature in the training model, receive the recommended temperature, which should be set in the air conditioner, and transfer the recommended temperature to an external device, wherein the trained training model performs training to provide the recommended temperature based on the set temperature and the current room ambient temperature. 23. The method of claim 22, further comprising the step of: receive information about the external environment of the air conditioner, wherein obtaining the recommended temperature to be set in the air conditioner comprises the step of entering the current ambient temperature in the room and the external environment information into the trained training model to obtain the recommended temperature to be set in the air conditioner. 24. The method according to p. 22, wherein storing the trained training model comprises storing a plurality of training models for each operating mode of the air conditioner, and wherein obtaining the recommended temperature to be set in the air conditioner comprises the step of entering the current ambient temperature in the room into the trained training model corresponding to the current operating mode of the air conditioner to obtain the recommended temperature to be set in the air conditioner. 25. The method of providing the recommended temperature of the air conditioner, while the mentioned method contains the steps at which: read the current ambient temperature in the air conditioner room, transfer the read current ambient temperature in the room to an external device, receiving the recommended temperature, which is the result obtained by obtaining the set temperature set in the air conditioner and the current ambient temperature in the air conditioner room at the time of setting the set temperature, apply the set temperature and the current room ambient temperature in the training model from the external device depending on the transmission of the current room ambient temperature, and get the recommended temperature set in the air conditioner, set the accepted recommended temperature in the air conditioner, wherein the training model is a training model trained using the set temperature preset in the air conditioner and the current room ambient temperature set, and wherein the training model performs training to provide the recommended temperature based on the set temperature and the current room ambient temperature. 26. A method for controlling a user terminal air conditioner, said method comprising the steps of: receiving a user input signal depending on the user input selecting the user interface (UI) of the artificial intelligence operation, transmitting a control command requesting the activation of the artificial intelligence mode of the air conditioner corresponding to the AI operation UI to the air conditioner, get the recommended temperature set in the air conditioner, which is the result obtained by: obtaining the set temperature set in the air conditioner and the current ambient temperature in the air conditioner room at the time of setting the set temperature, applying the set temperature and the current ambient temperature in the air conditioner room in the training model depending on the control command requesting the activation of the artificial intelligence mode of the air conditioner, and obtain the recommended temperature set in the air conditioner, and display the received recommended temperature on the screen, wherein the training model performs training to provide the recommended temperature based on the set temperature and the current room ambient temperature. 27. The method of claim 26, further comprising the step of: display the set temperature preset in the air conditioner as equal to the current ambient temperature in the room, together with the recommended temperature. 28. A method for generating a training model of a network system including an air conditioner and a training model server, said method comprising the steps of: receiving, by means of the air conditioner, a user control signal adjusting the temperature, transmitting by means of the air conditioner the set temperature and the current ambient temperature in the air conditioner room to an external device, generating, by means of the training model server, a training model using the set temperature and the current room ambient temperature, and storing by the training model server the generated training model to provide the recommended temperature of the air conditioner, at the same time, the formation of a training model by means of a server of training models using a given temperature and the current ambient temperature in the room contains: obtaining the set temperature set in the air conditioner and the current ambient temperature in the air conditioner room during the setting of the set temperature, applying the set temperature and the current room temperature of the air conditioner in the training model, and obtaining the recommended temperature set in the air conditioner. 29. A method for providing a recommended temperature in a network system including an air conditioner and a training model server, said method comprising the steps of: using the air conditioner, transmit the current ambient temperature in the air conditioner room to an external device, obtain, through the training model server, the recommended temperature of the air conditioner by applying the set temperature and the current room ambient temperature in the training model, and transmit the recommended temperature to an external device via the air conditioner, at the same time, obtaining the recommended temperature of the air conditioner by means of the training model server by applying the set temperature and the current ambient temperature in the room in the training model contains: obtaining the set temperature set in the air conditioner and the current ambient temperature in the air conditioner room during the setting of the set temperature, applying the set temperature and the current room temperature of the air conditioner in the training model, and obtaining the recommended temperature set in the air conditioner. 30. A method for controlling an air conditioner in a network system including an air conditioner and a user terminal, said method comprising the steps of: receiving, by means of the user terminal, a user input signal depending on the user input selecting the user interface (UI) of the artificial intelligence operation, transmitting, by the user terminal, a control command requesting the activation of the artificial intelligence mode of the air conditioner corresponding to the AI operation UI to the air conditioner, transmitting by means of the air conditioner the set temperature of the air conditioner and the current ambient temperature in the air conditioner room to the external device in response to receiving a control command requesting the activation of the artificial intelligence mode of the air conditioner, receiving by the air conditioner the recommended temperature, which is the result obtained by applying the set temperature and the current room ambient temperature in the training model, from an external device depending on the transmission of the set temperature and the current room ambient temperature, and adjust the accepted recommended temperature in the air conditioner by means of the air conditioner, wherein the training model is a training model trained using a plurality of set temperatures pre-set in the air conditioner and a plurality of current room ambient temperatures, wherein the training model performs training to provide the recommended temperature based on the set temperature and the current room ambient temperature.

Images