CN110986306A

CN110986306A - Method, device and equipment for adjusting room temperature based on machine learning and storage medium

Info

Publication number: CN110986306A
Application number: CN201911208143.0A
Authority: CN
Inventors: 温晗; 姜昊琦; 秦国祥; 邢戈; 杨华
Original assignee: Shanghai Envision Innovation Intelligent Technology Co Ltd; Envision Digital International Pte Ltd
Current assignee: Shanghai Envision Innovation Intelligent Technology Co Ltd; Envision Digital International Pte Ltd
Priority date: 2019-11-30
Filing date: 2019-11-30
Publication date: 2020-04-10
Anticipated expiration: 2039-11-30
Also published as: CN110986306B

Abstract

The application discloses a room temperature adjusting method, device, equipment and storage medium based on machine learning, and belongs to the field of computers. The method comprises the following steps: acquiring regional information and outdoor information of the temperature controller, wherein the regional information is used for indicating a region where a target room is located, and the outdoor information is at least used for representing outdoor temperature and humidity information of the target room; determining a target temperature prediction model from a plurality of temperature prediction models according to regional information of a target room, wherein the target temperature prediction model is a model obtained by training according to historical use records of all temperature controllers in the region, and the historical use records comprise sample data consisting of a plurality of groups of outdoor information and indoor temperatures set by a user; extracting an outdoor feature vector from outdoor information of a target room; calling a target temperature prediction model to predict the outdoor characteristic vector to obtain the indoor target temperature of the target room; and adjusting the actual temperature of the indoor area where the temperature controller is located according to the indoor target temperature.

Description

Method, device and equipment for adjusting room temperature based on machine learning and storage medium

Technical Field

The present application relates to the field of sensors, and in particular, to a method, an apparatus, a device, and a storage medium for room temperature adjustment based on machine learning.

Background

Fan Coil (FCU) is a commonly used end device for air conditioning systems, and a Fan Coil controls the temperature of a room by sending cooled or heated air into the room through a cold or hot water Coil in the Unit. The fan coil temperature controller is a control device used for controlling water flow circulation in a cold water coil and a hot water coil in the fan coil, and the fan coil is controlled by the temperature controller to ensure that the indoor temperature meets the requirements of users.

In the correlation technique, the exhaust port position of fan coil is provided with temperature sensor, regards temperature sensor sensing fan coil's air supply temperature as indoor temperature, compares the air supply temperature that senses with the temperature of settlement, and the rivers circulation mode among fan coil temperature controller control cold water coil and the hot water coil according to the comparative result to adjust fan coil's air supply temperature, make indoor temperature use the settlement temperature as the benchmark, undulant at certain within range.

Based on the above situation, the fan coil temperature controller adjusts the air supply temperature or the air volume based on the temperature sensed by the temperature sensor, so that the fan coil temperature controller is not accurate enough when adjusting the indoor temperature.

Disclosure of Invention

The embodiment of the application provides a room temperature adjusting method, a room temperature adjusting device, room temperature adjusting equipment and a storage medium based on machine learning, and can solve the problem that a fan coil temperature controller in the related art is not accurate enough when the room temperature is adjusted. The technical scheme is as follows:

according to one aspect of the present application, there is provided a room temperature adjusting method based on machine learning, the method including:

acquiring regional information and outdoor information of a temperature controller, wherein the regional information is used for indicating a region where a target room is located, and the outdoor information is at least used for representing outdoor temperature and humidity information of the target room;

determining a target temperature prediction model from a plurality of temperature prediction models according to the regional information of the target room, wherein the target temperature prediction model is trained according to historical usage records of all temperature controllers in the region, and the historical usage records comprise sample data consisting of a plurality of groups of outdoor information and indoor temperatures set by users;

extracting an outdoor feature vector from outdoor information of the target room;

calling the target temperature prediction model to predict the outdoor characteristic vector to obtain the indoor target temperature of the target room;

and adjusting the actual temperature of the indoor area where the temperature controller is located according to the indoor target temperature.

According to another aspect of the present application, there is provided a temperature regulation system, the system comprising: the server, the temperature controller and the air supply device; the server is connected with the temperature controller through a network, and the temperature controller is connected with the air supply device through a network or a circuit;

the server is used for acquiring regional information and outdoor information of the temperature controller, the regional information is used for indicating a region where a target room is located, and the outdoor information is at least used for representing outdoor temperature and humidity information of the target room;

the server is used for determining a target temperature prediction model from a plurality of temperature prediction models according to the regional information of the target room, wherein the target temperature prediction model is a model obtained by training according to historical usage records of all temperature controllers in the region, and the historical usage records comprise sample data consisting of a plurality of groups of outdoor information and indoor temperatures set by users;

the server is used for extracting an outdoor feature vector from outdoor information of the target room;

the server is used for calling the target temperature prediction model to predict the outdoor characteristic vector to obtain the indoor target temperature of the target room;

the temperature controller is used for adjusting the actual temperature of the indoor area where the temperature controller is located according to the indoor target temperature;

and the air supply device is used for carrying out air supply operation on the indoor area according to the difference value between the indoor target temperature and the actual temperature of the indoor area.

According to another aspect of the present application, there is provided a machine learning-based room temperature adjusting apparatus, the apparatus including:

the temperature controller comprises an acquisition module, a storage module and a control module, wherein the acquisition module is used for acquiring regional information and outdoor information of the temperature controller, the regional information is used for indicating a region where a target room is located, and the outdoor information is at least used for representing outdoor temperature and humidity information of the target room;

the processing module is used for determining a target temperature prediction model from a plurality of temperature prediction models according to the regional information of the target room, the target temperature prediction model is trained according to historical usage records of all temperature controllers in the region, and the historical usage records comprise sample data consisting of a plurality of groups of outdoor information and indoor temperatures set by a user;

an extraction module for extracting an outdoor feature vector from outdoor information of the target room;

the target temperature prediction model is used for predicting the outdoor characteristic vector to obtain the indoor target temperature of the target room;

and the adjusting module is used for adjusting the actual temperature of the indoor area where the temperature controller is located according to the indoor target temperature.

According to another aspect of the present application, there is provided a computer apparatus provided with a sensor, the computer apparatus including: a processor and a memory having stored therein at least one instruction, at least one program, a set of codes, or a set of instructions that is loaded and executed by the processor to implement a machine learning based room temperature adjustment method as described above.

According to another aspect of the present application, there is provided a computer-readable storage medium having stored therein at least one instruction, at least one program, code set, or set of instructions that is loaded and executed by the processor to implement the machine-learning based room temperature adjustment method as described above.

The beneficial effects brought by the technical scheme provided by the embodiment of the application at least comprise:

the target temperature prediction model is determined by obtaining the regional information and the outdoor information of the temperature controller, the target temperature prediction model is called to predict the outdoor characteristic vector extracted from the outdoor information to obtain the indoor target temperature of a target room, the actual temperature of the indoor region is adjusted according to the predicted target temperature, and the indoor target temperature is predicted through the outdoor information, so that the temperature controller is more accurate in adjusting the actual temperature of the indoor region.

Drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.

FIG. 1 is a schematic illustration of an implementation environment provided by an exemplary embodiment of the present application;

FIG. 2 is a block diagram of a server provided in an exemplary embodiment of the present application;

FIG. 3 is a schematic block diagram of a temperature regulation system provided in an exemplary embodiment of the present application;

FIG. 4 is a flow chart of a room temperature adjustment method based on machine learning provided by an exemplary embodiment of the present application;

FIG. 5 is a flow chart of a room temperature adjustment method based on machine learning provided by another exemplary embodiment of the present application;

fig. 6 is a flowchart illustrating a method for acquiring outdoor information of a thermostat according to an exemplary embodiment of the present application;

FIG. 7 is a schematic diagram of outdoor information provided by an exemplary embodiment of the present application;

fig. 8 is a flowchart illustrating a method for acquiring outdoor information of a thermostat according to another exemplary embodiment of the present application;

FIG. 9 is a flow chart of a method of training a temperature prediction model provided by an exemplary embodiment of the present application;

FIG. 10 is a schematic diagram of a thermostat provided in an exemplary embodiment of the present application;

FIG. 11 is a flow chart illustrating a method for a thermostat to adjust a room temperature according to an exemplary embodiment of the present application;

fig. 12 is a flowchart of a method for adjusting a room temperature by using a thermostat according to another exemplary embodiment of the present application;

FIG. 13 is a block diagram of an apparatus for machine learning based room temperature adjustment provided by an exemplary embodiment of the present application;

fig. 14 is a schematic device structure diagram of a computer apparatus according to an exemplary embodiment of the present application.

Detailed Description

To make the objects, technical solutions and advantages of the present application more clear, embodiments of the present application will be described in further detail below with reference to the accompanying drawings.

First, terms referred to in the embodiments of the present application are described:

fan Coil (Fan Coil Unit, FCU): the air conditioner is a common terminal device of an air conditioning system, and a fan coil sends cooled or heated air into a room through a cold water coil or a hot water coil in a unit to control the temperature in the room. The fan coil is usually composed of components such as a heat exchanger, a water pipe, a filter, a fan, a water receiving pipe, an exhaust valve and a bracket, and the fan coil can be divided into the following components according to the assembly form: horizontal concealed installation, horizontal exposed installation, vertical concealed installation, vertical exposed installation and clamping installation. The two-pipe fan coil adopts different water flow circulation modes in different seasons, for example, in summer, cold water circulates in the coil to play a refrigeration role; in winter, hot water circulates in the coil pipe to play a role in heating. In some commercial parks, such as shopping malls and office buildings, four-pipe fan coils are adopted, cold water and hot water can be simultaneously circulated in the coils, and different temperature adjustments can be carried out according to the requirements of different rooms or places. Illustratively, store A requires a temperature of 23 ℃ and store B requires a temperature of 30 ℃.

Temperature controller: the temperature controller in this application embodiment refers to the temperature controller of control fan coil, also is named fan coil temperature controller, and the user accessible temperature controller adjusts the temperature in the region that fan coil is located. The fan coil temperature controller is an independent closed-loop temperature regulating system and consists of a temperature sensor, a double-level controller, a temperature setting mechanism, a manual switch and a cold-hot switching device. The temperature controller compares the indoor temperature measured by the temperature sensor with the set temperature, and sends out a two-position control signal according to the temperature difference, the cold-hot switching device controls the water valves of the cold water circulation pipeline and the hot water circulation pipeline according to the two-position control signal, and the temperature of the fan coil supplying air to the indoor is adjusted in a mode of switching the water flow circulation in the coil, so that the indoor temperature fluctuates in a certain range by taking the set temperature as the reference.

FIG. 1 illustrates a block diagram of an environment in which a temperature regulation system provided by an exemplary embodiment of the present application may be implemented. The implementation environment comprises: a thermostat 130, a server cluster 140, and a fan coil 170.

The thermostat 130 is connected to the server cluster 140 through a network 150, the thermostat 130 and the fan coil 170 are connected through a network or circuit, and the thermostat 130 may be installed on a wall of an indoor area. Optionally, a temperature sensor for measuring a temperature of the indoor area and a distance detector for measuring whether there is a movable object, such as a person or a raised pet, in the indoor area are provided in the thermostat 130. Optionally, the thermostat 130 is installed and operated with an application program supporting temperature adjustment and an application program supporting connection, the application program supporting temperature adjustment enables the thermostat 130 to control the fan coil 170 to adjust the temperature of the indoor area, and the application program supporting connection is used for matching and connecting the thermostat 130 with the fan coil 170. Optionally, the temperature controller 130 is provided with physical keys, or the temperature controller 130 is provided with a display screen, or the temperature controller 130 is provided with physical keys and a display screen, and the temperature controller 130 may also be a terminal used by a user, such as at least one of a mobile phone, a laptop, a desktop computer, and a tablet computer.

Optionally, an account is registered in the application, and the account has authority to control the fan coil.

The server cluster 140 includes at least one of a server, a plurality of servers, a cloud computing platform, and a virtualization center. The server cluster 140 is used to provide background services for applications with augmented reality functionality. Optionally, the server cluster 140 undertakes primary computing work, and the thermostat 130 undertakes secondary computing work; or, the server cluster 140 undertakes the secondary computing work, and the thermostat 130 undertakes the primary computing work; or, the server cluster 140 and the thermostat 130 perform cooperative computing by using a distributed computing architecture.

Optionally, the server cluster 140 includes: the system comprises an access server and a background server. The access server is used for providing access service and information transceiving service of the thermostat 130 and forwarding the effective information to the background server. The background server is used for providing background services of the application program, such as: at least one of an external temperature information service, a user information service, and a thermostat's operation status service, optionally, the thermostat 130 may acquire information related to the outdoor temperature from the server cluster 140. The background server can be one or more. When the background servers are multiple, at least one of the following forms exists: at least two background servers exist for providing different services, and at least two background servers exist for providing the same service, which is not limited in the embodiment of the present application.

Optionally, the thermostat 130 is connected to the server cluster 140 via a first network 160 and the thermostat 130 is connected to the fan coil 170 via a second network 180. Alternatively, the first network connection 160 may be a metropolitan area network, a local area network, a fiber network, etc. in a wired network, or a mobile communication network (e.g., at least one of 2G, or 3G, or 4G, or 5G) or a Wireless Fidelity (WiFi) network in a Wireless network. Alternatively, the second network 180 may be the same or different type of connection as the first network 160, or a different network within the same network type. Illustratively, the temperature controller 130 is connected with the fan coil 170 through a wireless network, such as at least one of ZigBee technology, bluetooth technology, and wireless USB technology.

Fig. 2 shows a schematic structural diagram of a server provided in an exemplary embodiment of the present application. The server may be the server shown in fig. 1. Specifically, the method comprises the following steps:

the server 200 includes a Central Processing Unit (CPU) 201, a system Memory 204 including a Random Access Memory (RAM) 202 and a Read Only Memory (ROM) 203, and a system bus 205 connecting the system Memory 204 and the Central Processing Unit 201. The server 200 also includes a basic input/output System (I/O System)206, which facilitates the transfer of information between devices within the computer, and a mass storage device 207 for storing an operating System 213, application programs 214, and other program modules 215.

The basic input/output system 206 includes a display 208 for displaying information and an input device 209, such as a mouse, keyboard, etc., for user input of information. Wherein a display 208 and an input device 209 are connected to the central processing unit 201 through an input output controller 210 connected to the system bus 205. The basic input/output system 206 may also include an input/output controller 210 for receiving and processing input from a number of other devices, such as a keyboard, mouse, or electronic stylus. Similarly, input-output controller 210 also provides output to a display screen, a printer, or other type of output device.

The mass storage device 207 is connected to the central processing unit 201 through a mass storage controller (not shown) connected to the system bus 205. The mass storage device 207 and its associated computer-readable media provide non-volatile storage for the server 200. That is, the mass storage device 207 may include a computer-readable medium (not shown) such as a hard disk or Compact disk Read Only Memory (CD-ROM) drive.

Without loss of generality, computer readable media may comprise computer storage media and communication media. Computer storage media includes volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data. Computer storage media includes RAM, ROM, Erasable Programmable Read-Only Memory (EPROM), Electrically Erasable Programmable Read-Only Memory (EEPROM), flash Memory or other solid state Memory technology, CD-ROM, Digital Versatile Disks (DVD), or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices. Of course, those skilled in the art will appreciate that computer storage media is not limited to the foregoing. The system memory 203 and mass storage device 207 described above may be collectively referred to as memory.

According to various embodiments of the present application, server 200 may also operate as a remote computer connected to a network through a network, such as the Internet. That is, the server 200 may be connected to the network 212 through the network interface unit 211 connected to the system bus 205, or the network interface unit 211 may be used to connect to other types of networks or remote computer systems (not shown).

The memory further includes one or more programs, and the one or more programs are stored in the memory and configured to be executed by the CPU.

Fig. 3 shows a schematic structural framework diagram of a temperature regulation system provided in an exemplary embodiment of the present application, where the temperature regulation system 30 includes a cloud server 302 and a thermostat 303, public or external information 301 is stored in the cloud server 302, and the thermostat 303 is connected to the cloud server 302 through a network. The public or external information 301 includes at least one of weather forecast data, season data, room location data, indoor personnel data, and outdoor temperature data, and the public/external information 301 information is uploaded by a user through the cloud server 302 or transmitted to the temperature regulation system 30 through a network connection by other cloud servers. Optionally, a user account is registered in the temperature controller, or a terminal device used by the user is registered with the user account, and the user account has a management authority for the temperature controller and an authority for controlling the fan coil to adjust the room temperature. Optionally, each thermostat 303 corresponds to one or more user accounts, and each account corresponds to one or more thermostats 303.

The operation of the user on the thermostat 303 is uploaded to the cloud server 302, and the operation of the user includes at least one operation of a temperature set by the user each time, a time corresponding to the set temperature, a time for turning on the thermostat 303, a time for turning off the thermostat 303, an operation state of the thermostat 303, an operation mode of the thermostat 303, or a temperature adjustment mode. Illustratively, the cloud server 302 processes the historical temperature setting of the user to generate a temperature profile that conforms to the usage habits of the user. The cloud server 302 may also monitor the operating state of the thermostat 303, and illustratively, when the temperature set by the thermostat 303 does not conform to the temperature curve, the cloud server 302 may send a prompt to the thermostat 303, or send a prompt to a terminal (such as a mobile phone) used by a user, and the user may directly set the thermostat 303, or set the thermostat 303 through the terminal.

The in-room system configuration 31 includes an in-room sensor network 310 and a fan coil 320. Optionally, the in-room sensor network 310 includes at least one of a sub-temperature and humidity sampling device, a network, and a temperature detection device. Alternatively, the temperature in the room may also be measured by a temperature sensor built into the thermostat 303.

The thermostat 303 controls the fan coil 320, optionally, the fan coil 320 includes an external communication module 321, a control unit 322, a storage unit 323, a microcontroller 324, a built-in sensor 325, and an ad hoc network module 326, wherein the storage unit 323 and the microcontroller 324 are optional modules. The external communication module 321 is used for connecting a network, so that the fan coil 320 can be connected with the temperature controller 303, the fan coil 320 can upload temperature data to the cloud server 302, or download temperature data, and the temperature data can be outdoor weather data or temperature data in a building in the same area; the control unit 322 is used for controlling an alternating current power supply, a gear mode of the fan, a cold water valve and a hot water valve of the coil pipe; the storage unit 323 is used for rapidly reading data and storing the data when the fan coil 320 is powered off; microcontroller 324 is used to receive remote inputs and upgrade fan coil 320 online; the built-in sensor 325 is used for detecting the indoor temperature of the controlled area and the movable object information of the controlled area; the ad hoc network module 326 is used to implement the gateway function of the fan coil 320, and local peripheral wireless devices can be accessed through the ad hoc network module 326.

Fig. 4 shows a flowchart of a room temperature adjustment method based on machine learning according to an exemplary embodiment of the present application. The method can be applied to the server cluster 140 shown in fig. 1 and also can be applied to the cloud server 302 shown in fig. 3, and the method comprises the following steps:

step 401, obtaining area information and outdoor information of the temperature controller, where the area information is used to indicate an area where a target room is located, and the outdoor information is at least used to represent outdoor temperature and humidity information of the target room.

Optionally, the regional information includes any one of a region, a community, a commercial park, an office building, and a landscape park. The outdoor information comprises at least one of weather type, outdoor temperature, outdoor humidity, the floor where the target room is located, the house type orientation of the target room, and the ecological environment of the area where the target room is located (such as the area where the target room is located near the lake).

Optionally, a thermostat is disposed in the target room, and the thermostat may be installed in the target room, or a user may carry the thermostat with him in the target room, and obtain the area where the target room is located according to the area information of the thermostat. Alternatively, the target room is the residence of any residential building in a community, or the target room is a store in a commercial campus, or the target room is an office at any floor of an office building. The outdoor temperature and humidity information corresponding to different target rooms is different, like a residential building in a community, outdoor temperature and humidity information corresponding to a residential building close to the street is different from that corresponding to a residential building close to the middle of the community, and outdoor temperature and humidity information corresponding to residences on different floors in the same residential building is different.

Optionally, the outdoor temperature and humidity information includes at least one of outdoor season, weather type (such as sunny day or cloudy day), air temperature (including lowest air temperature, highest air temperature, and air temperature-time variation curve), humidity (including maximum humidity, minimum humidity, and humidity-time variation curve), dressing index (recommending that the user wears clothes corresponding to weather), and geographic location of the target room.

Alternatively, the zone information of the thermostat may be acquired first and then the outdoor information of the thermostat may be acquired, or the outdoor information of the thermostat may be acquired first and then the zone information of the thermostat may be acquired, or both the zone information and the outdoor information of the thermostat may be acquired.

Step 402, according to the regional information of the target room, determining a target temperature prediction model from a plurality of temperature prediction models, wherein the target temperature prediction model is a model trained according to historical usage records of all temperature controllers in the region, and the historical usage records comprise sample data consisting of a plurality of groups of outdoor information and indoor temperatures set by a user.

Optionally, the plurality of temperature prediction models are prediction models divided according to regions, such as a temperature prediction model a corresponding to the cell a and a temperature prediction model B corresponding to the cell B. Optionally, the plurality of temperature prediction models are prediction models divided according to the floor where the target room is located, such as a temperature prediction model c corresponding to the tenth floor of an office building and a temperature prediction model d corresponding to the fifteenth floor.

In one example, the target temperature prediction model is a model trained according to historical usage records of temperature controllers corresponding to all rooms in a floor, and if an office floor includes ten rooms and each room is provided with a temperature controller, the target temperature prediction model is trained according to the historical usage records of the ten temperature controllers. In one example, the historical usage record includes sample data consisting of a plurality of sets of outdoor information, indoor temperature set by the user, time and frequency at which the user sets the indoor temperature.

Step 403, extracting an outdoor feature vector from the outdoor information of the target room.

Optionally, an outdoor feature vector satisfying a correlation condition with the set indoor temperature is extracted from the outdoor information, the outdoor feature vector being of at least one dimension. In one example, the outdoor feature vector extracted from the outdoor information includes a house type of the target room, an area of the target room, an orientation of the target room, a floor on which the target room is located, and an outdoor ecological environment corresponding to the target room. Optionally, the feature vector is extracted by a target temperature prediction model, or other model, or server.

And step 404, calling a target temperature prediction model to predict the outdoor characteristic vector to obtain the indoor target temperature of the target room.

Alternatively, the target temperature prediction model may be one or more models, and when the target temperature prediction model is a plurality of models, it may be the same model, or different models of the same type, or different types of models. In one example, the target temperature prediction models are two models, the target temperature prediction models are a first target temperature prediction model divided according to regions and a second target temperature prediction model divided according to floors, the outdoor characteristic vector comprises the region where the target room is located and the floor where the target room is located, the first target temperature prediction model is called to predict the outdoor characteristic vector to obtain the range where the indoor target temperature is located, and then the second target temperature prediction model is called to predict the outdoor characteristic vector to obtain the indoor target temperature of the target room.

Step 405, adjusting the actual temperature of the indoor area where the temperature controller is located according to the indoor target temperature.

Optionally, the temperature controller is controlled according to a difference between the target indoor temperature and an actual indoor temperature, and the temperature controller controls the fan coil to adjust the actual indoor temperature. Optionally, the user may also manually adjust the indoor target temperature according to the target temperature prediction model.

In summary, in the method provided in this embodiment, the target temperature prediction model is determined by obtaining the area information and the outdoor information of the temperature controller, the target temperature prediction model is called to predict the outdoor eigenvector extracted from the outdoor information, so as to obtain the indoor target temperature of the target room, the actual temperature of the indoor area is adjusted according to the predicted target temperature, and the indoor target temperature is predicted by the outdoor information, so that the temperature controller is more accurate in adjusting the actual temperature of the indoor area.

Fig. 5 is a flowchart illustrating a method for machine learning based room temperature adjustment according to another exemplary embodiment of the present application. The method can be applied to the server cluster 140 shown in fig. 1 and also can be applied to the cloud server 302 shown in fig. 3, and the method comprises the following steps:

step 501a, obtaining an internet protocol address of the temperature controller.

Optionally, each thermostat is provided with an internet protocol address (IP), where the IP refers to an address provided by an internet protocol, and a logical address is allocated to each network and each host on the internet, where the logical address has a mapping relationship with an actual physical address of a user.

Step 502a, determining a room address of the target room according to the internet protocol address.

Optionally, a mapping relationship exists between the internet protocol and the room address of the target room, and the room address of the target room is determined according to the mapping relationship. The IP address may deviate from the actual target room address, and the room address of the target room may be obtained by combining a Positioning technology, such as Global Positioning System (GPS).

In step 503a, the zone information is determined according to the room address.

Optionally, the geographical location, room type, house type, floor and other regional information of the target room can be determined according to the room address. In one example, the room address of the target room is a, a corresponds to business campus a, the target room is located in office building D in business campus a, and at the tenth floor of office building D, the target room is located in the north area of the tenth floor of office building D.

And step 501b, acquiring installation record information of the temperature controller during installation.

Optionally, there is an installation log when the thermostat is installed in the target room. The installation record information may be acquired from a server for managing installation information of the thermostat, or may be acquired from a property management server corresponding to an installation room. In one example, the property management server stores the installation record of the temperature controller in the office building D, and the property management server is a server of the property management center corresponding to the business park a.

Step 502b, determining the room address of the target room according to the installation record information.

By installing the recorded information, it can be determined which rooms the thermostat is installed in the office building D, and the room address of the target room is obtained from these rooms.

Step 503b, determining the zone information according to the room address.

Step 503b is in principle consistent with step 503a and will not be described here.

It is understood that steps 501a to 503a, and steps 501b and 503b may be implemented individually or in combination. Optionally, the obtaining of the area information of the target room includes at least two methods, and when there is another method for obtaining the area information of the target room, the corresponding method may be implemented alone, or may be implemented in any combination with the above methods.

Optionally, the area information of the thermostat is obtained, and meanwhile, the corresponding outdoor information of the thermostat is obtained.

Step 504, according to the regional information of the target room, determining a target temperature prediction model from a plurality of temperature prediction models, wherein the target temperature prediction model is a model trained according to historical usage records of all temperature controllers in the region, and the historical usage records comprise sample data consisting of a plurality of groups of outdoor information and indoor temperatures set by users.

Step 505, an outdoor feature vector is extracted from the outdoor information of the target room.

Step 506, calling a target temperature prediction model to predict the outdoor characteristic vector to obtain the indoor target temperature of the target room.

The principle of steps 504 to 506 is the same as that of steps 402 to 404 shown in fig. 4, and the details are not repeated here.

And step 507, adjusting the actual temperature of the indoor area where the temperature controller is located according to the indoor target temperature.

Alternatively, the adjustment may be made according to a difference between the indoor target temperature and the actual temperature of the indoor area. In one example, adjusting the actual temperature of the indoor area further comprises the steps of:

at step 5071, the actual temperature of the indoor area is obtained.

At step 5072, a difference between the indoor target temperature and the actual temperature of the indoor area is obtained.

At step 5073, a target gear is determined from the at least two gears based on the difference.

In one example, the thermostat is provided with three stages, a low stage, a middle stage, and a high stage, and the difference between the indoor target temperature and the actual temperature of the indoor area is 1 ℃, it is determined that the target stage is the low stage.

Step 5074, the adjustment mode controls the temperature adjustment system to deliver an air volume to the indoor area, the air volume for adjusting an actual temperature of the indoor area.

Illustratively, the air volume corresponding to the low gear is 100 cubic meters per hour, the air volume corresponding to the medium gear is 200 cubic meters per hour, and the air volume corresponding to the high gear is 300 cubic meters per hour.

In summary, in the method provided in this embodiment, the actual address of the target room is determined by the IP address, the installation record information, and the like, and the area information of the temperature controller in the target room is obtained according to the actual address of the target room, so that the outdoor information corresponding to the target room can be determined, and the temperature prediction model is invoked to predict the outdoor feature vector according to the outdoor information, so as to obtain the indoor temperature of the target room, so that the temperature controller can adjust the indoor temperature more accurately.

Fig. 6 is a flowchart illustrating a method for acquiring outdoor information of a thermostat, where the method may be applied to the server cluster 140 shown in fig. 1 or the cloud server 302 shown in fig. 3, and the method includes the following steps:

step 601, acquiring an IP address of the temperature controller.

And step 602, determining the room address of the target room according to the IP address.

Step 601 and step 602 are consistent with the principle of step 501a and step 502a, and are not described herein again.

Step 603, obtaining the corresponding relation between the area information and the outdoor information of the temperature controller from the weather forecast information source according to the room address.

The area information of the temperature controller is used for indicating the area of the target room, and the area information comprises any one of areas, communities, commercial parks, office buildings and garden scenic spots. The outdoor information is used for representing outdoor temperature and humidity information of the target room and position information of the target room, and the outdoor information comprises at least one of weather type, outdoor temperature, outdoor humidity, a floor where the target room is located and house type orientation of the target room. When the room address of the target room is determined, outdoor information about the target room may be determined.

The weather forecast information source is an information source for providing weather conditions of each region, and the weather forecast information source comprises a server for recording weather information by a weather bureau, or weather information of a region where a target room is located is acquired by a weather satellite, or a platform for each weather forecast (such as a website of the weather bureau or an application program for the weather forecast).

And step 604, determining the outdoor information according to the corresponding relation.

The relationship between the IP addresses of the pair of thermostats and the outdoor information of the room address of the target room will be described with reference to the table.

Watch 1

The information corresponding to the target room, such as at least one of the orientation, the house type, and the area of the target room, can be obtained according to the room address of the target room.

Taking the area where the target room is located as a cell as an example, the relationship between the area information and the outdoor information of the thermostat will be described with reference to fig. 7.

As shown in fig. 7, the residential building 11 and the residential building 12 are two residential buildings in the same residential area, and on the basis of 106 solar rays, the residential building 11 is located in front of the residential building 12, and the residential building 11 blocks a part of the sunlight irradiated into the residential building 12. The temperature corresponding to rooms on different floors in the same residential building is different, and the temperature corresponding to rooms on the same floor but in different directions in the same residential building is different. Illustratively, the room 101 and the room 103 in the residential building 11 are located on different floors in the same residential building, the room 101 is closer to the top floor of the residential building 11, and the lighting is better, and the room 103 is closer to the bottom floor of the residential building 11, if there are other buildings in front of the residential building 11, the lighting of the room 103 is worse than that of the room 101, and therefore, the indoor temperature corresponding to the room 101 is higher, and the indoor temperature corresponding to the room 103 is lower. Illustratively, the room 101 and the room 102 are respectively located in different directions on the same floor in the same residential building, the room 101 is located in a direction toward the sun, and the room 102 is located in a direction away from the sun, so that the room 101 corresponds to a higher indoor temperature and the room 102 corresponds to a lower indoor temperature. Illustratively, the room 104 is located in the residential building 12 and is located in a direction facing away from the sun, and the room 104 has a lower indoor temperature than the room 102. The room 105 is located on the side of the residential building, and the room is less exposed to sunlight, and the room temperature corresponding to the room is lower than the room temperatures corresponding to other rooms.

Optionally, the temperature information and the humidity information of the residential building 11 and the residential building 12 in a period of time are collected, and a corresponding temperature database is established according to the temperature information and the humidity information. In one example, the temperature database includes temperature information and humidity information for the room 101 over the course of one year, such as the following for the room 101 over the entire day of 10 months and 1 day of 2019: the maximum temperature was 26 ℃, the minimum temperature was 19 ℃, the maximum humidity was 40%, the minimum humidity was 30%, and the sun exposure time was 8 hours. Illustratively, the temperature database may establish a corresponding indoor temperature for the room 101 at each moment in time. The characteristic vector can be extracted from the temperature information and the humidity information corresponding to the room, and a target temperature prediction model is called to predict the characteristic vector to obtain the indoor target temperature of the target room.

In the following, another method for acquiring outdoor information of a thermostat is described, and fig. 8 shows a flowchart of a method for acquiring outdoor information of a thermostat according to another exemplary embodiment of the present application, where the method may be applied to the server cluster 140 shown in fig. 1 and may also be applied to the cloud server 302 shown in fig. 3, and the method includes the following steps:

step 801, obtaining installation record information of the temperature controller during installation.

Step 802, determining the room address of the target room according to the installation record information.

In one example, the property center of the district A is provided with a server, the server stores installation records of the temperature controllers, a house type graph included in the district A and a residential building distribution diagram of the district A, the residential building where the temperature controllers are located can be determined according to the installation records of the temperature controllers, and outdoor information such as house types, areas, orientations and the like of target rooms where the temperature controllers are located can be determined according to the distribution diagram of the residential building of the district A. Optionally, the residential building where the temperature controller is located and the floor number where the temperature controller is located can be obtained according to the address of the target room, and outdoor information such as the house type, the area, the orientation and the like of the target room can be determined according to the residential building distribution diagram of the A cell and the house type diagram.

And 803, acquiring the corresponding relation between the area information and the outdoor information of the temperature controller from the weather forecast information source according to the room address.

Step 803 is consistent with the principle of step 603 shown in fig. 6, and is not described here again.

And step 804, determining the outdoor information according to the corresponding relation.

The relationship between the installation record information of the two pairs of temperature controllers and the outdoor information of the room address of the target room will be described.

Watch two

The following describes a method of training the temperature prediction model. FIG. 9 illustrates a method for training a temperature prediction model provided by an exemplary embodiment of the present application. The method can be applied to the server cluster 140 shown in fig. 1 and also can be applied to the cloud server 302 shown in fig. 3, and the method comprises the following steps:

step 901, obtaining historical usage records of each temperature controller in the area, wherein the historical usage records comprise sample data formed by a plurality of groups of outdoor information and indoor temperatures set by users.

Optionally, the historical usage record comprises at least one of the following records: the set temperature (i.e., target temperature) of the thermostat, the range of temperature change, the frequency of use of the thermostat, the most commonly used target temperature, and the length of time the target temperature lasts. Optionally, the outdoor information may be obtained from servers of various organizations, for example, the server provided in the embodiment of the present application may obtain the temperature data from a server for managing weather information, where the server for managing weather information may be a server corresponding to a weather application program, and may also be a server corresponding to a weather bureau. Optionally, the indoor temperature set by the user may be obtained according to a user account registered in the thermostat, or obtained through a terminal used by the user, in which a user account for controlling the thermostat is registered.

Step 902, for each group of sample data, extracting the outdoor information to obtain a sample feature vector.

Optionally, the extracted feature vector is an at least one-dimensional feature vector. A feature vector as extracted from a set of sample data includes the following five dimensions: weather type, air temperature, room address of a target room, indoor temperature set by a user, and frequency of the same temperature set by the user.

And 903, inputting the sample characteristic vector into a temperature prediction model to obtain a predicted indoor temperature.

In step 904, an error between the predicted indoor temperature and the user-set indoor temperature is calculated.

Alternatively, any error function may be used to calculate the error between the predicted indoor temperature and the indoor temperature set by the user, such as an euclidean loss function.

Step 905, training a temperature prediction model by adopting an error back propagation algorithm according to the error.

And training the temperature prediction model through an error back propagation algorithm according to the error loss obtained in the step 904 to obtain the trained temperature prediction model.

Fig. 10 shows a schematic structural diagram of a thermostat provided in an exemplary embodiment of the present application. The temperature controller is connected with a cloud server 1001 through a wireless network-system level chip module 1002, and can acquire regional information and outdoor information of the temperature controller from the cloud server 1001. Optionally, the thermostat is provided with a plurality of interfaces, wherein the serial peripheral interface 1005a is used for connecting the flexible display screen 1101, the serial peripheral interface 1005b is used for connecting the flash memory chip 1014, the flash memory chip 1014 is a non-volatile storage device, when the device fails or is powered off, data stored in the flash memory chip 1014 cannot be lost, the integrated circuit bus 1006 is used for connecting the temperature and humidity sensor 1012, the universal one-step transceiver transmitter 1009 is used for connecting the peak voltage module 1013, the relay 1007 is used for connecting the fan control 1015, the cold water valve 1016 and the hot water valve 1017, and the electric energy conversion module 1010 is used for connecting the power module 1003 and is connected with the 220 v power voltage 1004 through the power module 1003. The flexible display screen 1011 and the purple peak module 1013 are optional modules. The fan control 1015 is used to control the gear change of the temperature controller, the gear corresponds to the air volume delivered into the room by the fan coil or the air supply device, and if the middle gear is switched to the low gear, the large air volume is adjusted to the small air volume by the fan coil or the air supply device.

The embodiment of the present application further provides a method for adjusting an actual temperature of an indoor area by using a thermostat, which is described below with reference to fig. 11 and 12. Fig. 11 is a flowchart illustrating a method for adjusting an actual temperature of an indoor area by a thermostat according to an exemplary embodiment of the present application, the method being applied to the thermostat shown in fig. 10, and the method including the steps of:

step 1101, start.

The temperature controller starts to work.

In step 1102a, it is determined whether the thermostat is in a cooling mode.

The cooling mode means that the water flow circulation in the fan coil is cold water, the fan coil delivers cold air to the indoor, and the temperature of the indoor environment is reduced. If the temperature controller is in the cooling mode, go to step 1103 a; otherwise, go to step 1102 b.

Step 1103a, determine if the thermostat is in automatic mode.

If the thermostat is in automatic mode, go to step 1105; otherwise, go to step 1104 a.

And step 1104a, executing for 5 minutes according to the current wind speed, and switching to an automatic mode.

This step allows the fan coil to maintain a stable temperature of the indoor environment before entering the automatic mode.

In step 1105, the current ambient temperature is obtained in the automatic control mode.

In step 1106, the set temperature is compared to the current temperature difference Δ T.

Step 1107a, determine whether the difference satisfies Δ T > 2.

If the difference between the set temperature and the current temperature is delta T > 2, entering step 1107 b; otherwise, go to step 1108 a.

Step 1107b, turn on the high range fan.

In step 1108a, it is determined whether the difference value is Δ T > 1.

If the difference between the set temperature and the current temperature is Δ T > 1, go to step 1108 b; otherwise, go to step 1109 a.

Step 1108b, turn on the mid-range fan.

And step 1109a, judging whether the difference value accords with delta T larger than 0.

If the difference between the set temperature and the current temperature is more than 0, entering step 1109 b; otherwise, go to step 1110.

And step 1109b, turning on a low-speed fan.

In step 1110, it is determined whether the difference is equal to Δ T equal to 0.

If the difference between the set temperature and the current temperature is equal to Δ T ═ 0, go to step 1111; otherwise, go to step 1107 a.

Step 1111, the fan is closed, the cold water valve is closed, and the hot water valve is closed.

Alternatively, the fan, the cold water valve and the hot water valve can be closed in a certain sequence or simultaneously.

In step 1102b, it is determined whether the thermostat is in a heating mode.

And step 1103b, judging whether the temperature controller is in the automatic mode.

If the thermostat is in automatic mode, go to step 1105; otherwise, go to step 1104 b.

And step 1104b, executing the wind speed for 5 minutes, and switching to an automatic mode.

It is understood that steps 1102b through 1104b are consistent with the principles of steps 1102a through 1104 a.

Fig. 12 is a flowchart illustrating a method of adjusting an actual temperature of an indoor area by a thermostat according to another exemplary embodiment of the present application, which is applied to the thermostat shown in fig. 10, and includes the following steps:

step 1201, start.

The indoor ambient temperature is regulated.

Step 1202, determine whether the indoor ambient temperature reaches a set temperature.

If the indoor environment temperature reaches the set temperature, go to step 1203; otherwise, return to step 1201.

At step 1203, a close command is generated.

Optionally, the shutdown command is used to shut down a fan, a cold water valve, and a hot water valve in a fan coil.

In step 1204, a counter starts counting time.

And step 1205, judging whether the stabilization time for the temperature to reach the set temperature reaches 60 seconds.

If the stabilization time reaches 60 seconds, go to step 1206 a; otherwise, go back to step 1204.

In step 1206a, it is determined whether the difference between the indoor temperature and the set temperature is greater than 0 and Δ T is less than or equal to 1.

If the difference between the set temperature and the current temperature is more than 0 and less than or equal to 1, the process goes to step 1207; otherwise, go to step 1206 b.

Step 1207, enter intelligent trim mode.

And step 1208, judging whether the indoor temperature reaches a preset temperature.

If the indoor temperature reaches the set temperature, go to step 1209; otherwise, return to step 1207 is entered.

Step 1209, the close command is validated.

Step 1210, close the fan, close the cold water valve, and close the hot water valve.

And step 1206b, judging whether the difference value between the indoor temperature and the set temperature is more than or equal to 1 and less than or equal to 2.

If the difference between the set temperature and the current temperature is equal to or greater than 1 and equal to or less than 2, entering step 1207 b; otherwise, go to step 1206 c.

In step 1207b, the fan is turned on mid-speed.

In step 1206c, it is determined whether the difference between the indoor temperature and the set temperature is Δ T > 2.

If the difference between the set temperature and the current temperature is delta T > 2, go to step 1207 c; otherwise, go to step 1211.

In step 1207c, the fan is turned on.

Step 1211, enter intelligent acceleration mode.

The indoor temperature is quickly reached to the set temperature through the intelligent acceleration mode.

In step 1212, it is determined whether the indoor temperature reaches a predetermined temperature.

If the indoor temperature reaches the set temperature, go to step 1203; otherwise, go to step 1213.

Step 1213, determine if the stabilization time after reaching the set temperature is greater than 900 seconds.

If the stabilization time is greater than 900 seconds, go to step 1207; otherwise, step 1214 is entered.

Step 1214, the intelligent acceleration mode is continuously maintained and timed.

Optionally, the above-mentioned indoor temperature accessible sub humiture collection equipment of acquireing obtains accurate indoor temperature, and if the indoor environment is provided with a plurality of sub humiture collection equipment, then indoor temperature is the average value of the temperature that a plurality of sub humiture collection equipment gathered. If the indoor environment has no temperature and humidity acquisition equipment, the indoor temperature is the temperature detected by a temperature sensor arranged in the temperature controller. Optionally, the sub-temperature and humidity sampling device may be a thermometer, a hygrometer, or other devices that can collect temperature and humidity data.

It can be understood that the unit of the temperature difference Δ T between the indoor temperature and the set temperature is celsius degree (° c), the temperature difference range, the stable time when the indoor temperature reaches the set temperature, and the execution time of the operating mode are selectable, and may be manually set by a user, set by a server, or default set of the temperature controller.

The embodiment of the application also provides a temperature regulating system, which comprises a server, a temperature controller and an air supply device, wherein the server is connected with the temperature controller through a network, and the temperature controller is connected with the air supply device through a network or a circuit.

Alternatively, the server is a cloud server 302 as shown in fig. 3, the thermostat is a thermostat 303 as shown in fig. 3, and the air blowing device is a fan coil 320 as shown in fig. 3.

The server is used for acquiring the area information and the outdoor information of the temperature controller, the area information is used for indicating the area where the target room is located, and the outdoor information is at least used for representing the outdoor temperature and humidity information of the target room; determining a target temperature prediction model from a plurality of temperature prediction models according to regional information of a target room, wherein the target temperature prediction model is a model obtained by training according to historical use records of all temperature controllers in the region, and the historical use records comprise sample data consisting of a plurality of groups of outdoor information and indoor temperatures set by a user; extracting an outdoor feature vector from outdoor information of the target room; and calling the target temperature prediction model to predict the outdoor characteristic vector to obtain the indoor target temperature of the target room.

And the temperature controller is used for adjusting the actual temperature of the indoor area where the temperature controller is located according to the indoor target temperature.

And the air supply device is used for carrying out air supply operation on the indoor area according to the difference value of the indoor target temperature and the actual temperature of the indoor area.

The following are embodiments of the apparatus of the present application, and for details that are not described in detail in the embodiments of the apparatus, reference may be made to corresponding descriptions in the above method embodiments, and details are not described herein again.

Fig. 13 shows a schematic structural diagram of a machine learning-based room temperature adjusting device according to an exemplary embodiment of the present application. The apparatus can be implemented as all or a part of a terminal by software, hardware or a combination of both, and includes: an acquisition module 1310, a processing module 1320, an extraction module 1330, a target temperature prediction model 1340, and an adjustment module 1350.

The obtaining module 1310 is configured to obtain area information and outdoor information of the temperature controller, where the area information is used to indicate an area where a target room is located, and the outdoor information is at least used to represent outdoor temperature and humidity information of the target room;

a processing module 1320, configured to determine a target temperature prediction model from multiple temperature prediction models according to the region information of the target room, where the target temperature prediction model is a model trained according to historical usage records of each temperature controller in the region, and the historical usage records include sample data composed of multiple sets of outdoor information and the indoor temperature set by the user;

an extracting module 1330, configured to extract an outdoor feature vector from outdoor information of the target room;

the target temperature prediction model 1340 is used for predicting the outdoor characteristic vector to obtain the indoor target temperature of the target room;

and an adjusting module 1350, configured to adjust an actual temperature of an indoor area where the temperature controller is located according to the indoor target temperature.

In an alternative embodiment, the obtaining module 1310 is configured to obtain an internet protocol IP address of the thermostat;

the processing module 1320, configured to determine a room address of the target room according to the IP address;

the processing module 1320 is configured to determine area information according to a room address, where the area information includes: any one of a region, a community, a business park, and an office building.

In an optional embodiment, the obtaining module 1310 is configured to obtain installation record information of the thermostat during installation;

the processing module 1320, configured to determine a room address of the target room according to the installation record information;

In an alternative embodiment, the apparatus further includes a computation module 1360 and a training module 1370;

the obtaining module 1310 is configured to obtain historical usage records of each temperature controller in an area, where the historical usage records include sample data composed of multiple sets of outdoor information and indoor temperatures set by a user;

the extracting module 1330 is configured to, for each group of sample data, extract the outdoor information to obtain a sample feature vector;

the processing module 1320 is configured to input the sample feature vector to the temperature prediction model to obtain a predicted indoor temperature;

the calculating module 1360 is used for calculating an error between the predicted indoor temperature and the indoor temperature set by the user;

the training module 1370 is configured to train the temperature prediction model by using an error back propagation algorithm according to the error.

In an alternative embodiment, the outdoor information includes: at least one of weather type, outdoor temperature, outdoor humidity, floor where the target room is located, and house type orientation of the target room.

In an optional embodiment, the obtaining module 1310 is configured to obtain an IP address of the thermostat;

the processing module 1320 is configured to obtain a corresponding relationship between the area information and the outdoor information of the thermostat from a weather forecast information source according to the room address;

the processing module 1320 is configured to determine the outdoor information according to the corresponding relationship.

In an alternative embodiment, the obtaining module 1310 is configured to obtain an actual temperature of an indoor area;

the obtaining module 1310 is configured to obtain a difference between an indoor target temperature and an actual temperature of an indoor area;

the processing module 1320 is configured to determine a target gear from at least two gears according to the difference;

the adjusting module 1350 is configured to adjust the mode to control the temperature adjusting system to deliver the air volume to the indoor area, where the air volume is used to adjust the actual temperature of the indoor area.

The present application also provides a computer device provided with a sensor 1403, illustratively, the sensor 1403 is a far and near infrared sensor, or the sensor 1403 is a temperature sensor. The computer device 1400 includes: a processor 1401 and a memory 1402, wherein the memory 1402 stores at least one instruction, at least one program, a set of codes, or a set of instructions, which are loaded and executed by the processor 1401 to implement the machine learning based room temperature adjusting method provided by the above-mentioned method embodiments.

The present application further provides a computer-readable storage medium, in which at least one instruction, at least one program, a code set, or a set of instructions is stored, and the at least one instruction, the at least one program, the code set, or the set of instructions is loaded and executed by a processor to implement the machine learning-based room temperature adjustment method provided by the above method embodiments.

It should be understood that reference to "a plurality" herein means two or more. "and/or" describes the association relationship of the associated objects, meaning that there may be three relationships, e.g., a and/or B, which may mean: a exists alone, A and B exist simultaneously, and B exists alone. The character "/" generally indicates that the former and latter associated objects are in an "or" relationship.

It will be understood by those skilled in the art that all or part of the steps for implementing the above embodiments may be implemented by hardware, or may be implemented by a program instructing relevant hardware, where the program may be stored in a computer-readable storage medium, and the above-mentioned storage medium may be a read-only memory, a magnetic disk or an optical disk, etc.

The above description is only exemplary of the present application and should not be taken as limiting, as any modification, equivalent replacement, or improvement made within the spirit and principle of the present application should be included in the protection scope of the present application.

Claims

1. A room temperature adjusting method based on machine learning, the method comprising:

2. The method of claim 1, wherein the obtaining zone information of the thermostat comprises:

acquiring an Internet Protocol (IP) address of the temperature controller; determining a room address of the target room according to the IP address; or acquiring installation record information of the temperature controller during installation; determining the room address of the target room according to the installation record information;

determining the zone information according to the room address, wherein the zone information comprises: any one of a region, a community, a business park, and an office building.

3. The method of any one of claims 1 to 3, wherein the temperature prediction model is trained by:

acquiring historical use records of all temperature controllers in the area, wherein the historical use records comprise sample data consisting of a plurality of groups of outdoor information and indoor temperatures set by a user;

for each group of the sample data, extracting the outdoor information to obtain a sample feature vector;

inputting the sample characteristic vector into the temperature prediction model to obtain a predicted indoor temperature;

calculating an error between the predicted indoor temperature and the user-set indoor temperature;

and training the temperature prediction model by adopting an error back propagation algorithm according to the error.

4. The method of any of claims 1 to 3, wherein the outdoor information comprises: at least one of a weather type, an outdoor temperature, an outdoor humidity, a floor on which the target room is located, and a house type orientation of the target room.

5. The method of claim 1, wherein the obtaining of the outdoor information of the thermostat comprises:

acquiring an IP address of the temperature controller; determining a room address of the target room according to the IP address; or acquiring installation record information of the temperature controller during installation; determining the room address of the target room according to the installation record information;

acquiring the corresponding relation between the area information of the temperature controller and the outdoor information from a weather forecast information source according to the room address;

and determining the outdoor information according to the corresponding relation.

6. The method according to any one of claims 1 to 3, wherein the adjusting the actual temperature of the indoor area where the thermostat is located according to the indoor target temperature comprises:

acquiring the actual temperature of the indoor area;

calculating a difference between the indoor target temperature and an actual temperature of the indoor area;

determining a target gear from at least two gears according to the difference;

and controlling a temperature adjusting system to convey air volume to the indoor area in an adjusting mode, wherein the air volume is used for adjusting the actual temperature of the indoor area.

7. A temperature conditioning system, characterized in that the system comprises: the server, the temperature controller and the air supply device; the server is connected with the temperature controller through a network, and the temperature controller is connected with the air supply device through a network or a circuit;

8. A machine learning based room temperature conditioning apparatus, the apparatus comprising:

9. A computer device having a sensor disposed thereon, the computer device comprising a processor and a memory, the memory having stored therein at least one instruction, at least one program, set of codes, or set of instructions, the at least one instruction, at least one program, set of codes, or set of instructions being loaded and executed by the processor to implement the machine learning based room temperature adjustment method of any of claims 1 to 6.

10. A computer-readable storage medium having stored therein at least one instruction, at least one program, a set of codes, or a set of instructions, which is loaded and executed by a processor to implement the machine learning based room temperature adjustment method of any one of claims 1 to 6.