WO2021179958A1 - 空调器的智能控制方法与空调器的智能控制设备 - Google Patents

空调器的智能控制方法与空调器的智能控制设备 Download PDF

Info

Publication number
WO2021179958A1
WO2021179958A1 PCT/CN2021/078687 CN2021078687W WO2021179958A1 WO 2021179958 A1 WO2021179958 A1 WO 2021179958A1 CN 2021078687 W CN2021078687 W CN 2021078687W WO 2021179958 A1 WO2021179958 A1 WO 2021179958A1
Authority
WO
WIPO (PCT)
Prior art keywords
air conditioner
data
record
records
user
Prior art date
Application number
PCT/CN2021/078687
Other languages
English (en)
French (fr)
Inventor
宋世芳
郭丽
吴丽琴
Original Assignee
青岛海尔空调器有限总公司
海尔智家股份有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 青岛海尔空调器有限总公司, 海尔智家股份有限公司 filed Critical 青岛海尔空调器有限总公司
Publication of WO2021179958A1 publication Critical patent/WO2021179958A1/zh

Links

Images

Classifications

    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F24—HEATING; RANGES; VENTILATING
    • F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
    • F24F11/00—Control or safety arrangements
    • F24F11/62—Control or safety arrangements characterised by the type of control or by internal processing, e.g. using fuzzy logic, adaptive control or estimation of values
    • F24F11/63—Electronic processing
    • F24F11/64—Electronic processing using pre-stored data
    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F24—HEATING; RANGES; VENTILATING
    • F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
    • F24F11/00—Control or safety arrangements
    • F24F11/50—Control or safety arrangements characterised by user interfaces or communication
    • F24F11/56—Remote control
    • F24F11/58—Remote control using Internet communication
    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F24—HEATING; RANGES; VENTILATING
    • F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
    • F24F11/00—Control or safety arrangements
    • F24F11/50—Control or safety arrangements characterised by user interfaces or communication
    • F24F11/61—Control or safety arrangements characterised by user interfaces or communication using timers
    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F24—HEATING; RANGES; VENTILATING
    • F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
    • F24F11/00—Control or safety arrangements
    • F24F11/62—Control or safety arrangements characterised by the type of control or by internal processing, e.g. using fuzzy logic, adaptive control or estimation of values
    • F24F11/63—Electronic processing
    • F24F11/65—Electronic processing for selecting an operating mode
    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F24—HEATING; RANGES; VENTILATING
    • F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
    • F24F2110/00—Control inputs relating to air properties
    • F24F2110/10—Temperature
    • F24F2110/12—Temperature of the outside air
    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F24—HEATING; RANGES; VENTILATING
    • F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
    • F24F2110/00—Control inputs relating to air properties
    • F24F2110/20—Humidity
    • F24F2110/22—Humidity of the outside air

Definitions

  • the invention relates to the control of smart home appliances, in particular to an intelligent control method of an air conditioner and an intelligent control device of the air conditioner.
  • An object of the present invention is to provide an intelligent control method of an air conditioner and an intelligent control device of an air conditioner that at least to some extent solve any of the technical problems in the above-mentioned related art.
  • a further object of the present invention is to realize the intelligent self-adjustment of the air conditioner according to the environmental data and improve the user experience.
  • Another further object of the present invention is to provide a personalized adjustment solution for users with unique usage habits.
  • the present invention provides an intelligent control method of an air conditioner, which includes: obtaining environmental data records of the area where the target controlled air conditioner is located within a set time period; Forecast, obtain a self-adjusting strategy; and control the target controlled air conditioner according to the self-adjusting strategy.
  • the step of obtaining environmental data records of the area where the target controlled air conditioner is located within a set time period includes: obtaining location information of the target controlled air conditioner, and obtaining outdoor environment records within the set time period according to the location information; And obtain the indoor environment record within the set time period uploaded by the target controlled air conditioner.
  • the self-adjustment strategy includes any one or more of the following: power-on time of the air conditioner, initial power-on mode of the air conditioner, initial power-on parameters of the air conditioner, and response actions of the air conditioner to changes in environmental detection data.
  • the training process of the self-adjusting model includes: subscribing to batches of air conditioner log data from the operating data platform, the operating data platform is used to collect and record the operating log of the air conditioner; the log data is parsed into structured data and stored It is a sample database; use the data in the sample database for machine learning model training to obtain a self-adjusting model.
  • the step of parsing log data into structured data includes: extracting data according to type tags according to log data; classifying and storing the extracted data to obtain a candidate sample data table; checking the data in the candidate sample data table Perform statistical screening to obtain structured data.
  • the method further includes: obtaining the operation record uploaded by the target controlled air conditioner; extracting the manual adjustment record from the operation record; and after the manual adjustment record exceeds the set number threshold , Use the operating records to establish the personality adjustment model of the target controlled air conditioner, so as to use the personality adjustment model to formulate the follow-up control strategy of the target controlled air conditioner.
  • the method further includes: obtaining the operation record uploaded by the target controlled air conditioner; extracting the manual adjustment record from the operation record; and after the manual adjustment record exceeds the set number threshold , Use the user behavior record of the target controlled air conditioner to establish a user portrait model of the target controlled air conditioner, so as to use the user portrait model to formulate the subsequent control strategy of the target controlled air conditioner.
  • the step of using the user behavior record of the target controlled air conditioner to establish a user portrait model of the target controlled air conditioner includes: obtaining the behavior record of the target user within a set time period, and the behavior record includes at least the air conditioner usage record; The behavior record and the operation record of the target controlled air conditioner are used as training samples, and the portrait model is obtained through training.
  • the behavior record further includes any one or more of the following: the user's travel record, the user's location record, and the user's physiological characteristic record.
  • an intelligent control device for an air conditioner which includes: a processor; and a memory, in which a control program is stored, and the control program is used to implement any of the above Intelligent control method of air conditioner.
  • the intelligent control method of the air conditioner of the present invention uses a machine learning algorithm to establish a self-adjusting model in advance, and uses the self-adjusting model to predict environmental data records to obtain a self-adjusting strategy for controlling the area where the air conditioner is located. Since the self-adjustment strategy is based on environmental data, it can accurately meet the user's comfort requirements for the environment and provide the user's experience.
  • the self-adjusting model used in the intelligent control method of the air conditioner of the present invention is obtained by training on the big data collected by the operating data platform. Due to the use of sufficient samples, the self-adjusting model is accurate and effective, and can meet the recommended requirements of the air conditioner control strategy.
  • the intelligent control method of the air conditioner of the present invention can also establish a personalized portrait model for users with unique usage records to meet the intelligent requirements of these users.
  • Fig. 1 is a schematic diagram of an environment where an air conditioner is located according to an embodiment of the present invention
  • Fig. 2 is a schematic diagram of data interaction of an air conditioner intelligent control system according to an embodiment of the present invention
  • Fig. 3 is a schematic diagram of an intelligent control device of an air conditioner according to an embodiment of the present invention.
  • Fig. 4 is a schematic diagram of an intelligent control method of an air conditioner according to an embodiment of the present invention.
  • Fig. 5 is a schematic diagram of training a self-adjusting model in an intelligent control method of an air conditioner according to an embodiment of the present invention.
  • Fig. 6 is a schematic flowchart of an intelligent control method for an air conditioner according to an embodiment of the present invention.
  • Fig. 1 is a schematic diagram of an environment where an air conditioner is located according to an embodiment of the present invention.
  • the air conditioner intelligent control system of this embodiment has multiple network data platforms.
  • Multiple network data platforms may include: an operation data platform 100 for collecting and recording air conditioner operation logs, an information collecting platform 200 for collecting and recording environmental information and/or user behavior information, and the like.
  • the operation data platform 100 can perform data interaction with the air conditioner 110, collect operation records of the air conditioner 110, and obtain massive air conditioner log data. These air conditioner log data can not only be used for after-sales service, status tracking, and fault diagnosis, but also can be used to provide sample data required for machine learning.
  • the information collection platform 200 is used to collect and record environmental information.
  • the environmental information may include not only indoor environmental information, but also outdoor environmental information.
  • the environmental information may include, but is not limited to: one or more of temperature, humidity, time, climate, region, and air pollutant concentration.
  • the information collection platform 200 can interact with various environmental detection equipment or environmental data to obtain the aforementioned environmental data.
  • the information collection platform 200 can be used to collect various behaviors of users. These behaviors may include, but are not limited to: records of the user's use of various household appliances (for example, records of adjusting air conditioners), and travel records of users (for example, driving Record), user's location record, user's physiological characteristic record (such as body temperature, heart rate, blood pressure), user's exercise record, user's work and rest record and so on.
  • the information collection platform 200 can interact with various devices of the user, such as interacting with mobile terminals, vehicles, smart home appliances, wearable devices, and so on.
  • the behavior records collected in the user information collection platform 200 provide users with a basis for various data analysis.
  • Fig. 2 is a schematic diagram of data interaction of an air conditioner intelligent control system according to an embodiment of the present invention.
  • the air conditioner 110 uploads the operation record to the operation data platform 100.
  • the operating data platform 100 forms a big data platform for the air conditioner 110 by collecting the operating records of the air conditioner 110.
  • the air conditioner intelligent control device 300 subscribes to batches of air conditioner log data from the operating data platform 100, and uses a machine learning algorithm to train to obtain a self-adjusting model.
  • the intelligent control device 300 of the air conditioner records the environmental data of the area where the target controlled air conditioner 110 is located in the user terminal 330 and other detection devices within a set time period, and enters the self-adjustment after sorting
  • the model after the data processing of the self-adjusting model, finally gets the self-adjusting strategy.
  • the smart air conditioner 110 adjusts the environment of the target user according to the self-adjustment strategy.
  • users can get a comfortable environment that meets their own requirements without any adjustment operations.
  • Fig. 3 is a schematic diagram of a smart control device 300 of an air conditioner according to an embodiment of the present invention.
  • the intelligent control device 300 may generally include: a memory 320 and a processor 310, wherein the memory 320 stores a control program 321, and when the control program 321 is executed by the processor 310, it is used to realize the intelligence of the air conditioner of this embodiment. Control Method.
  • the processor 310 may be a central processing unit (central processing unit, CPU for short), or a digital processing unit, and so on.
  • the processor 310 transmits and receives data through a communication interface.
  • the memory 320 is used to store a program executed by the processor 310.
  • the memory 320 is any medium that can be used to carry or store desired program codes in the form of instructions or data structures and that can be accessed by a computer, and may also be a combination of multiple memories.
  • the above-mentioned control program 321 can be downloaded from a computer-readable storage medium to a corresponding computing/processing device or downloaded and installed to the intelligent control device 300 via a network (for example, the Internet, a local area network, a wide area network, and/or a wireless network).
  • Fig. 4 is a schematic diagram of an intelligent control method of an air conditioner according to an embodiment of the present invention.
  • the intelligent control method of the air conditioner may generally include:
  • Step S402 Acquire environmental data records of the area where the target controlled air conditioner is located within a set time period; the environmental data records not only do not include indoor environmental data, but may also include outdoor environmental data.
  • These environmental data records reflect the operating characteristics of the target controlled air conditioner. For example, when acquiring outdoor environment data, the location information of the target controlled air conditioner can be acquired, and the outdoor environment records within a set time period can be acquired according to the location information. These outdoor environment records can be obtained by obtaining detection data from detection equipment, or by obtaining data collected by other data collection platforms. For another example, when acquiring indoor environment data, acquire indoor environment records within a set time period uploaded by the target controlled air conditioner.
  • the air conditioner can update the latest control strategy in time, it is possible to choose to use environmental data records in a set time period, for example, environmental data records in the last month or several weeks.
  • the pre-trained self-adjusting model is used to predict the environmental data records to obtain a self-adjusting strategy.
  • the self-adjustment strategy may include any one or more of the following: the switch-on time of the air conditioner, the initial startup mode of the air conditioner, the initial startup parameters of the air conditioner, and the response action of the air conditioner to changes in environmental detection data.
  • the self-adjustment strategy is not only used to output target parameters (for example, target temperature), but can also include adjustment actions of target controlled air conditioners for different situations. For example, when there is a different temperature difference between the actual temperature and the target temperature, different temperature adjustment methods of the target controlled air conditioner; for example, in the sleep mode, the control curve of the target controlled air conditioner, and so on.
  • step S406 the target controlled air conditioner is controlled according to the self-adjusting strategy.
  • the target controlled air conditioner can start by itself, determine the target parameters, and perform response adjustments without user intervention, which greatly improves the user's comfort and user experience.
  • a machine learning algorithm is used to establish a self-adjusting model in advance, and the self-adjusting model is used to predict environmental data records to obtain a self-adjusting strategy for controlling the area where the air conditioner is located. Since the self-adjustment strategy is based on environmental data, it can accurately meet the user's comfort requirements for the environment and provide the user's experience.
  • Fig. 5 is a schematic diagram of training a self-adjusting model in an intelligent control method of an air conditioner according to an embodiment of the present invention.
  • the training process of the self-adjusting model may include:
  • Step S502 subscribing to batches of air conditioner log data from the operation data platform, and the operation data platform is used to collect and record the operation log of the air conditioner.
  • the log data is parsed into structured data and stored as a sample database.
  • the analysis process of structured data can include: extracting data according to type labels according to log data; classifying and storing the extracted data to obtain a candidate sample data table; statistically filtering the data in the candidate sample data table to obtain a structured data data.
  • the air conditioner can be saved in categories such as on time, off time, initial temperature, target temperature, wind speed, wind direction, operating mode, compressor frequency, adaptation scenario, target parameter adjustment time, etc., for use in subsequent training.
  • Step S506 Use the data in the sample database to train the machine learning model to obtain a self-adjusting model.
  • the machine learning model may be able to learn certain knowledge and capabilities from existing data for processing new data, and may be designed to perform various tasks. In this embodiment, it is used to determine the control strategy of the air conditioner.
  • machine learning models include, but are not limited to, various types of deep neural networks (DNN), support vector machines (SVM), decision trees, random forest models, and so on.
  • machine learning models can also be called "learning networks".
  • the neural network control model can use various known network structures suitable for supervised learning, such as basic neural network structures such as perceptron models, classifier models, and Hopfield networks.
  • Various corresponding mainstream training methods can also be used The determination of model parameters in this embodiment.
  • Example machine learning models include neural networks or other multilayer nonlinear models.
  • Example neural networks include feedforward neural networks, deep neural networks, recurrent neural networks, and convolutional neural networks.
  • the machine learning model can be included in the server computing system (network data platform) or stored and implemented by the server computing system (network data platform) in other ways.
  • the server computing system (network data platform) communicates with the terminal device according to the client-server relationship .
  • the machine learning model can be implemented as a part of a web service by a server computing system (network data platform). Therefore, one or more models can be stored and implemented at the terminal device and/or one or more models can be stored and implemented at a server computing system (network data platform).
  • the server computing system may include one or more server computing devices or be implemented by one or more server computing devices in other ways.
  • server computing devices may operate according to a sequential computing architecture, a parallel computing architecture, or some combination thereof.
  • the terminal device (air conditioner and other user equipment) and/or the server computing system (network data platform) can train the model via interaction with the training computing system communicatively coupled through the network.
  • the training computing system can be separated from the server computing system (network data platform), or can be a part of the server computing system (network data platform).
  • the terminal equipment (air conditioner and other user equipment) and the server computing system network can interact through any type of communication network, such as a local area network (such as an intranet), a wide area network (such as the Internet), or some combination thereof, and can include any number Wired or wireless link.
  • communication through the network can be via any type of wired and/or wireless connection, using various communication protocols (for example, TCP/IP, HTTP, SMTP, FTP), encoding or format (for example, HTML, XML), and/or Protection scheme (for example, VPN, secure HTTP, SSL) to carry.
  • the method of this embodiment also takes into account that some users have unique usage habits, and there may be a self-adjustment strategy obtained by predicting environmental data records through a self-adjusting model, which cannot meet the comfort requirements of these users.
  • a personality adjustment model can also be established based on manual adjustment records.
  • the step of controlling the air conditioner in the environment of the target user may also include: obtaining the operation record uploaded by the target controlled air conditioner; extracting the manual adjustment record from the operation record; After a certain number of thresholds, the operation record is used to establish the personality adjustment model of the target controlled air conditioner, so as to use the personality adjustment model to formulate the subsequent control strategy of the target controlled air conditioner. That is, the operation record uploaded by the target controlled air conditioner is used to establish a personality adjustment model, and the target controlled air conditioner is controlled in a targeted manner.
  • the above-mentioned set threshold is used to determine whether the manual adjustment record of the target user can meet the quantity requirement as the sample data.
  • the step of controlling the air conditioner in the environment where the target user is located according to the self-adjustment strategy it may further include: obtaining manual adjustment records uploaded by the air conditioner in the environment where the target user is located; judging whether the manual adjustment records of the target user exceed the setting Threshold of times; if yes, obtain the target user's behavior record and manual adjustment record, and use the target user's behavior record and manual adjustment record as training samples to train the target user's portrait model. That is to use the target user's behavior records and manual adjustment records to establish a portrait model.
  • After training the portrait model of the target user it also includes: using the portrait model to determine the control strategy of the target user; and controlling the air conditioner in the environment where the target user is located according to the control strategy.
  • the above-mentioned set threshold is used to determine whether the manual adjustment record of the target user can meet the quantity requirement as the sample data.
  • the portrait model of the target user can be used as a dedicated model of the user to meet the intelligence requirements of these users.
  • the training method of the portrait model can also adopt the current mainstream training method.
  • the behavior record can be preprocessed, for example, according to a predetermined correlation table to filter out related behaviors from the behavior record, where the correlation table is used to store various user behaviors and air conditioning The relevance of the running state of the device; the use of related behaviors to train the portrait model. Since there are many types of behavior records obtained, there are differences in the data that different users may obtain. These behavior record data may be related to the use of air conditioners, and some are less related to the use of air conditioners. Therefore, in this embodiment, the data correlation between various user behaviors and air conditioner adjustment and environmental comfort requirements is calculated in advance to obtain a correlation table.
  • the behavior records whose relevance reaches the preset threshold are selected from the behavior records according to the correlation correspondence table, which are used as the training samples of the portrait model of the target user.
  • the calculation of the above-mentioned data correlation can be performed by using the prior art of statistical calculation in the prior art. After screening, the data format of behavior records can also be unified, facilitating sample data processing and other processing.
  • the behavior record also includes any one or more of the following: the user's travel record, the user's location record, and the user's physiological characteristic record.
  • the record of the user adjusting the air conditioner directly reflects the user's control habits; the user's travel record can determine the time when the user arrives at the home or work place, so as to determine the opening time of the corresponding air conditioner; the user's location record can be used to determine the environment Data, climate, etc.; the user's physiological characteristics record reflects the user's physical state, which also directly affects the comfort requirements.
  • the scalability of various records of the user is fully considered, and various behavior information of the user can be fully collected.
  • Different data collection devices for users can collect different behavioral information, and select required relevant information through relevance screening.
  • Fig. 6 is a schematic flowchart of an intelligent control method for an air conditioner according to an embodiment of the present invention, and the process may include:
  • step S602 the log data of the air conditioner is counted by the running data platform, and the self-adjusting model is obtained through training;
  • step S604 the log data of the air conditioner is used as a sample, and a self-adjusting model is obtained through training;
  • Step S606 Obtain environmental data records of the area where the target controlled air conditioner is located within a set time period
  • Step S608 the environmental data record is input into the self-adjusting model, and the control strategy is determined
  • Step S610 using the control strategy to control the target controlled air conditioner
  • Step S612 whether the user manually adjusts the target controlled air conditioner, if no adjustment is made or the amount of adjustment data is small, continue to use the self-adjusting strategy to control the air conditioner;
  • step S614 if the user manually adjusts the air conditioner, it is determined whether the manual adjustment data meets the sample data requirements;
  • Step S616 After manually adjusting whether the data meets the sample data requirements, select user behavior information according to the correlation with the air conditioner adjustment;
  • step S618 machine learning model training is performed on the user behavior information to obtain a user portrait model
  • step S620 the subsequent control strategy of the target controlled air conditioner is formulated by using the user portrait model.
  • the environmental data records including indoor and outdoor temperature, indoor and outdoor humidity, region, time, day of the week, whether it is working day, etc. can be input into the self-adjusting model for prediction, and the self-adjusting strategy can be obtained .
  • the output parameters of the self-adjusting model may include: various startup parameters and their corresponding influence weights.
  • the final output parameters are: power-on time, season, month, whether it is working day, setting mode, indoor temperature, adjustment time interval, suitable temperature difference between indoor and outdoor.
  • a personalized portrait model for control you can, for example, record the user’s home time record, the user’s residential location record, the user’s transport use record, and the user’s air conditioner use record, combined with information such as seasons, workdays, and regional climate.
  • the training sample of the user profile model can be selected from the previous week or several weeks of data.
  • a machine learning algorithm is used to establish a self-adjusting model in advance, and the self-adjusting model is used to predict environmental data records to obtain a self-adjusting strategy for controlling the area where the air conditioner is located. Since the self-adjustment strategy is based on environmental data, it can accurately meet the user's comfort requirements for the environment and provide the user's experience.
  • the method of this embodiment can also establish a personalized portrait model for users with unique usage records to meet the intelligence requirements of these users.

Landscapes

  • Engineering & Computer Science (AREA)
  • Signal Processing (AREA)
  • Chemical & Material Sciences (AREA)
  • Combustion & Propulsion (AREA)
  • Mechanical Engineering (AREA)
  • General Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Fuzzy Systems (AREA)
  • Mathematical Physics (AREA)
  • Human Computer Interaction (AREA)
  • Air Conditioning Control Device (AREA)

Abstract

一种空调器的智能控制方法与空调器的智能控制设备。其中空调器的智能控制方法，包括：获取目标受控空调器所在区域在设定时间段内的环境数据记录(S402)；利用预先训练的自调整模型对环境数据记录进行预测，得到自调整策略(S404)；以及按照自调整策略对目标受控空调器进行控制(S406)。

Description

空调器的智能控制方法与空调器的智能控制设备 技术领域
本发明涉及智能家电控制，特别是涉及空调器的智能控制方法与空调器的智能控制设备。
背景技术
随着生活水平的日益提高，消费者对家电的选择不再是单单注重产品的质量，而是更注重产品能够带来的使用体验。
对于空调器之类的环境调节设备，用户的需求在于获得高舒适性的环境体验。为了满足用户的需求，空调器的功能逐渐扩展，控制也更加精细化。因此，空调器的使用也越来越复杂。现有技术中已经使用终端App来进行控制，然而这也使得用户学习使用的门槛越来越高，操作也越来越复杂。这反而给用户带来的不便。针对这一问题，现有技术并没有提供能够有效的解决方案。
发明内容
本发明的一个目的是要提供一种至少在一定程度上解决上述相关技术中的技术问题任一方面的空调器的智能控制方法与空调器的智能控制设备。
本发明一个进一步的目的是要空调器根据环境数据实现智能自调整，提高用户使用感受。
本发明另一个进一步的目的是要为使用习惯独特的用户提供个性化调整方案。
特别地，本发明提供了一种空调器的智能控制方法，其包括：获取目标受控空调器所在区域在设定时间段内的环境数据记录；利用预先训练的自调整模型对环境数据记录进行预测，得到自调整策略；以及按照自调整策略对目标受控空调器进行控制。
可选地，获取目标受控空调器所在区域在设定时间段内的环境数据记录的步骤包括：获取目标受控空调器的位置信息，根据位置信息获取设定时间段内的室外环境记录；以及获取目标受控空调器上传的设定时间段内的室内环境记录。
可选地，自调整策略包括以下任意一项或多项：空调器的开关机时刻、 空调器的开机初始模式、空调器的开机初始参数、空调器对环境检测数据变化的响应动作。
可选地，自调整模型的训练过程包括：从运行数据平台中订阅批量的空调器日志数据，运行数据平台用于采集并记录空调器的运行日志；将日志数据解析为结构化数据，并存储为样本数据库；使用样本数据库中的数据进行机器学习模型训练，得到自调整模型。
可选地，将日志数据解析为结构化数据的步骤包括：按照日志数据按照类型标签提取数据；将提取出的数据进行分类保存，得到备选样本数据表；对备选样本数据表中的数据进行统计筛选，得到结构化数据。
可选地，在按照自调整策略对空调器进行控制的步骤之后还包括：获取目标受控空调器上传的运行记录；从运行记录中提取手动调整记录；在手动调整记录超过设定次数阈值后，利用运行记录建立目标受控空调器的个性调整模型，以利用个性调整模型制定目标受控空调器的后续控制策略。
可选地，在按照自调整策略对空调器进行控制的步骤之后还包括：获取目标受控空调器上传的运行记录；从运行记录中提取手动调整记录；在手动调整记录超过设定次数阈值后，利用目标受控空调器的用户行为记录建立目标受控空调器的用户画像模型，以利用用户画像模型制定目标受控空调器的后续控制策略。
可选地，利用目标受控空调器的用户行为记录建立目标受控空调器的用户画像模型的步骤包括：获取目标用户在设定时间段内的行为记录，行为记录至少包括空调器使用记录；将行为记录以及目标受控空调器的运行记录作为训练样本，训练得到画像模型。
可选地，行为记录还包括以下任意一项或多项：用户的出行记录、用户的位置记录、用户的生理特征记录。
根据本发明的另一个方面，还提供了一种空调器的智能控制设备，其包括：处理器；以及存储器，存储器内存储有控制程序，控制程序被处理器执行时用于实现上述任一种空调器的智能控制方法。
本发明的空调器的智能控制方法，预先利用机器学习算法建立自调整模型，利用自调整模型对环境数据记录进行预测，得到用于对空调器所在区域进行控制的自调整策略。由于自调整策略基于环境数据得出，可以准确地满足用户对环境的舒适性要求，提供了用户的使用体验。
进一步地，本发明的空调器的智能控制方法中使用的自调整模型通过运行数据平台采集的大数据训练得出。由于采用的足够的样本，自调整模型准确有效，能够满足空调器控制策略的推荐要求。
更进一步地，本发明的空调器的智能控制方法，还可以对于使用记录比较独特的用户，建立个性化的画像模型，满足这些用户的智能要求。
根据下文结合附图对本发明具体实施例的详细描述，本领域技术人员将会更加明了本发明的上述以及其他目的、优点和特征。
附图说明
后文将参照附图以示例性而非限制性的方式详细描述本发明的一些具体实施例。附图中相同的附图标记标示了相同或类似的部件或部分。本领域技术人员应该理解，这些附图未必是按比例绘制的。附图中：
图1是根据本发明一个实施例的空调器所在环境的示意图；
图2是根据本发明一个实施例的空调器智能控制系统的数据交互示意图；
图3是根据本发明一个实施例的空调器的智能控制设备的示意图；
图4是根据本发明一个实施例的空调器的智能控制方法的示意图；
图5是根据本发明一个实施例的空调器的智能控制方法中训练自调整模型的示意图；以及
图6是根据本发明一个实施例的空调器的智能控制方法的流程示意图。
具体实施方式
图1是根据本发明一个实施例的空调器所在环境的示意图。本实施例的空调器智能控制系统具有多个网络数据平台。多个网络数据平台可以包括：用于采集并记录空调器运行日志的运行数据平台100、用于采集并记录环境信息和/或用户行为信息的信息采集平台200等。
运行数据平台100可以与空调器110进行数据交互，采集空调器110的运行记录，从而得到海量的空调器日志数据。这些空调器日志数据除了可以用于售后服务、状态跟踪、故障诊断之外，还可以用于提供机器学习所需的样本数据。
信息采集平台200一方面用于采集并记录环境信息，环境信息不仅可以包括室内环境信息，还可以包括室外环境信息。这些环境信息可以包括但不 限于：温度、湿度、时间、气候、地域、空气污染物浓度中的一种或多种。信息采集平台200可以各种环境检测设备或者环境数据等进行交互，以获取上述环境数据。
信息采集平台200另一方面用可以用于采集用户的各种行为，这些行为可以包括但不限于：用户使用各项家电设备的记录(例如调节空调器的记录)、用户的出行记录(例如行车记录)、用户的位置记录、用户的生理特征记录(例如体温、心率、血压)、用户的运动记录、用户的作息记录等等。信息采集平台200可以与用户的各项设备进行数据交互，例如与移动终端、交通工具、智能家电、可穿戴设备等进行交互。用户信息采集平台200内收集的行为记录，为用户提供了各项数据分析的基础。
由于对于信息技术领域的技术人员而言，上述网络数据平台的数据采集和存储技术本身为其所习知的技术，因此在此对数据采集和存储的具体流程不做赘述。
图2是根据本发明一个实施例的空调器智能控制系统的数据交互示意图。空调器110向运行数据平台100上传运行记录。运行数据平台100通过采集空调器110的运行记录，形成空调器110的大数据平台。空调器的智能控制设备300从运行数据平台100中订阅批量的空调器日志数据，并采用机器学习算法，训练得到自调整模型。
在空调器的智能运行状态下，空调器的智能控制设备300从用户终端330以及其他检测设备中目标受控空调器110所在区域在设定时间段内的环境数据记录，经整理后输入自调整模型，经自调整模型的数据处理，最终得到自调整策略。智能空调器110按照自调整策略对目标用户的所在环境进行调节。在智能运行状态下，用户无需进行任何调节操作，就可以得到符合自身要求的舒适环境。
图3是根据本发明一个实施例的空调器的智能控制设备300的示意图。该智能控制设备300可以包括一般性地可以包括：存储器320以及处理器310，其中存储器320内存储有控制程序321，控制程序321被处理器310执行时用于实现本实施例的空调器的智能控制方法。处理器310可以是一个中央处理单元(central processing unit，简称CPU)，或者为数字处理单元等等。处理器310通过通信接口收发数据。存储器320用于存储处理器310执行的程序。存储器320是能够用于携带或存储具有指令或数据结构形式的期望的 程序代码并能够由计算机存取的任何介质，也可以是多个存储器的组合。上述控制程序321可以从计算机可读存储介质下载到相应计算/处理设备或者经由网络(例如因特网、局域网、广域网和/或无线网络)下载并安装到智能控制设备300。
图4是根据本发明一个实施例的空调器的智能控制方法的示意图，该空调器的智能控制方法一般性地可以包括：
步骤S402，获取目标受控空调器所在区域在设定时间段内的环境数据记录；环境数据记录不仅仅不包括室内环境数据，还可以包括室外环境数据。这些环境数据记录体现了目标受控空调器的运行特征。例如在获取室外环境数据时，可以获取目标受控空调器的位置信息，根据位置信息获取设定时间段内的室外环境记录。这些室外环境记录可以通过获取检测设备的检测数据来得到，也可以通过获取其他数据采集平台收集的数据。又例如在获取室内环境数据时，获取目标受控空调器上传的设定时间段内的室内环境记录。
本实施例的方法，考虑到使得空调器能够及时更新最新的控制策略，因此可以选择使用设定时间段内的环境数据记录，例如最近一月或者数周的环境数据记录。
步骤S404，利用预先训练的自调整模型对环境数据记录进行预测，得到自调整策略。自调整策略可以包括以下任意一项或多项：空调器的开关机时刻、空调器的开机初始模式、空调器的开机初始参数、空调器对环境检测数据变化的响应动作。自调整策略并不仅仅用于输出目标参数(例如目标温度)，还可以包括针对不同情况目标受控空调器的调整动作。例如在实际温度与目标温度存在不同的温差下，目标受控空调器的不同调温方式；又例如在睡眠模式下，目标受控空调器的控制曲线等等。
步骤S406，按照自调整策略对目标受控空调器进行控制。控制过程中，目标受控空调器可以自行启动，确定目标参数，进行响应调整，无需要用户进行干预，大大提高了用户的舒适性以及使用体验。
本实施例的空调器的智能控制方法，预先利用机器学习算法建立自调整模型，利用自调整模型对环境数据记录进行预测，得到用于对空调器所在区域进行控制的自调整策略。由于自调整策略基于环境数据得出，可以准确地满足用户对环境的舒适性要求，提供了用户的使用体验。
图5是根据本发明一个实施例的空调器的智能控制方法中训练自调整模 型的示意图，该自调整模型的训练过程可以包括：
步骤S502，从运行数据平台中订阅批量的空调器日志数据，运行数据平台用于采集并记录空调器的运行日志。
步骤S504，将日志数据解析为结构化数据，并存储为样本数据库。结构化数据的解析过程可以包括：按照日志数据按照类型标签提取数据；将提取出的数据进行分类保存，得到备选样本数据表；对备选样本数据表中的数据进行统计筛选，得到结构化数据。例如可以分别针对空调器的开机时间、关机时间、初始温度、目标温度、风速、风向、运行模式、压缩机频率、适配场景、目标参数调整时间等分类进行保存，以便在后续训练时使用。
步骤S506，使用样本数据库中的数据进行机器学习模型训练，得到自调整模型。
机器学习模型可以是能够从已有数据中学习到一定的知识和能力用于处理新数据，并可以被设计用于执行各种任务，在本实施例中用于对空调器控制策略的确定。机器学习模型的示例包括但不限于各类深度神经网络(DNN)、支持向量机(SVM)、决策树、随机森林模型等等。在本领域内，机器学习模型也可以被称为“学习网络”。其中神经网络控制模型可以采用各种已知的适合有监督学习的网络结构，例如感知器模型，分类器模型，Hopfield网络等基本的神经网络结构，各种相应的主流训练方法也都可以用于本实施例的模型参数的确定。示例机器学习模型包括神经网络或其他多层非线性模型。示例神经网络包括前馈神经网络、深度神经网络、递归神经网络和卷积神经网络。
机器学习模型可以包括在服务器计算系统(网络数据平台)中或以其他方式由服务器计算系统(网络数据平台)存储和实现，服务器计算系统(网络数据平台)根据客户端-服务器关系与终端设备通信。例如，机器学习模型可以由服务器计算系统(网络数据平台)实现为web服务的一部分。因此，可以在终端设备处存储和实现一个或多个模型和/或可以在服务器计算系统(网络数据平台)处存储和实现一个或多个模型。
服务器计算系统(网络数据平台)可以包括一个或多个服务器计算设备或以其他方式由一个或多个服务器计算设备实现。在服务器计算系统包括多个服务器计算设备的情况下，这样的服务器计算设备可以根据顺序计算架构、并行计算架构或其一些组合来操作。
终端设备(空调器以及其他用户设备)和/或服务器计算系统(网络数据平台)可以经由与通过网络通信地耦接的训练计算系统的交互来训练模型。训练计算系统可以与服务器计算系统(网络数据平台)分离，或者可以是服务器计算系统(网络数据平台)的一部分。
终端设备(空调器以及其他用户设备)和服务器计算系统网络之间可以通过任何类型的通信网络进行交互，诸如局域网(例如内联网)、广域网(例如因特网)或其一些组合，并且可以包括任何数量的有线或无线链路。通常，通过网络的通信可以经由任何类型的有线和/或无线连接，使用各种通信协议(例如，TCP/IP、HTTP、SMTP、FTP)、编码或格式(例如、HTML、XML)和/或保护方案(例如、VPN、安全HTTP、SSL)来承载。
本实施例的方法还考虑到部分用户的使用习惯独特，可能出现通过自调整模型对环境数据记录进行预测得到的自调整策略，无法满足这些用户的舒适性要求。本实施例还可以根据手动调整记录来建立个性调整模型。
例如，在按照自调整策略对目标用户所在环境的空调器进行控制的步骤之后还可以包括：获取目标受控空调器上传的运行记录；从运行记录中提取手动调整记录；在手动调整记录超过设定次数阈值后，利用运行记录建立目标受控空调器的个性调整模型，以利用个性调整模型制定目标受控空调器的后续控制策略。也即利用目标受控空调器上传的运行记录建立个性调整模型，针对性地针对目标受控空调器进行控制。上述设定次数阈值用于判断目标用户的手动调整记录是否能够满足作为样本数据的数量要求。
又例如，在按照自调整策略对目标用户所在环境的空调器进行控制的步骤之后还可以包括：获取目标用户所在环境的空调器上传的手动调整记录；判断目标用户的手动调整记录是否超过设定次数阈值；若是，获取目标用户的行为记录以及手动调整记录，并将目标用户的行为记录以及手动调整记录作为训练样本，训练得到目标用户的画像模型。也即利用目标用户的行为记录以及手动调整记录建立画像模型。
在训练得到目标用户的画像模型之后还包括：使用画像模型确定目标用户的控制策略；以及按照控制策略对目标用户所在环境的空调器进行控制。上述设定次数阈值用于判断目标用户的手动调整记录是否能够满足作为样本数据的数量要求。
目标用户的画像模型可以作为该用户的专用模型，满足这些用户的智能 要求。画像模型的训练方法同样可以采用目前主流的训练方法。
在获取到用户的行为记录后，可以对行为记录进行预处理，例如根据预先确定的相关性对应表从行为记录中筛选出相关行为，其中相关性对应表用于保存有各类用户行为与空调器运行状态的相关性；利用相关行为训练画像模型。由于获取的行为记录的种类较多，对于不同用户可能获取到的数据也存在差别。这些行为记录数据可能与空调器的使用存在相关性，有些与空调器的使用相关性较低。因此本实施例预先对用户各项行为与空调器调节以及环境舒适性要求的数据相关性进行计算，得到相关性对应表。在获取到用户的行为记录后，根据相关性对应表从行为记录中筛选出相关性达到预设阈值的记录，作为目标用户的画像模型的训练样本。上述数据相关性进行计算可以采用现有技术中统计计算的现有技术进行。经过筛选，行为记录的数据格式也可以得到统一，便于样本数据的处理以及其他处理。
行为记录还包括以下任意一项或多项：用户的出行记录、用户的位置记录、用户的生理特征记录。其中用户调节空调器的记录直接反映了用户的调控习惯；用户的出行记录可以确定用户达到家居场所或者工作场所的时间，从而可以确定对应空调器的开启时间；用户的位置记录可以用于确定环境数据、气候等；用户的生理特征记录反映了用户的身体状态，这也直接影响了舒适性要求。
在本实施例的方法中，充分考虑了用户各种记录的扩展性，可以充分收集用户的各项行为信息。针对用户不同的数据收集装置，可以收集不同的行为信息，通过相关性的筛选，挑选需要的相关信息。
图6是根据本发明一个实施例的空调器的智能控制方法的流程示意图，该流程可以包括：
步骤S602，运行数据平台统计空调器日志数据，训练得到自调整模型；
步骤S604，将空调器日志数据作为样本，训练得到自调整模型；
步骤S606，获取目标受控空调器所在区域在设定时间段内的环境数据记录；
步骤S608，环境数据记录输入自调整模型，确定控制策略；
步骤S610，使用控制策略对目标受控空调器进行控制；
步骤S612，用户是否对目标受控空调器手工进行调整，如果未进行调整或调整数据量较小，则持续使用自调整策略对空调器进行控制；
步骤S614，若用户对空调器手工调整，则判断手工调整数据是否达到样本数据要求；
步骤S616，手工调整数据是否达到样本数据要求后，根据与空调器调节的相关性选取用户行为信息；
步骤S618，对用户行为信息进行机器学习模型训练，得到用户画像模型；
步骤S620，利用用户画像模型制定目标受控空调器的后续控制策略。
以确定空调器的开机参数为例进行介绍，可以将包含室内外温度、室内外湿度、地域、时间、周几、是否工作日等信息的环境数据记录输入自调整模型进行预测，得到自调整策略。自调整模型的输出参数可以包括：各项开机参数及其对应的影响权重。例如最终输出的参数为：开机时刻、季节、月份、是否工作日、设定模式、室内温度、调整时间间隔、室内外适宜温度差。
在利用个性化的画像模型进行控制时，可以例如将用户的到家时间记录、用户居住位置记录、用户的交通工具使用记录、用户使用空调器记录，结合季节、工作日信息、地域气候等信息作为用户画像模型的训练样本。上述记录可以选择此前一周或者数周的数据，在进行控制时，将近期用户的行为信息输入用户画像模型得到后续控制策略。
本实施例的空调器的智能控制方法，预先利用机器学习算法建立自调整模型，利用自调整模型对环境数据记录进行预测，得到用于对空调器所在区域进行控制的自调整策略。由于自调整策略基于环境数据得出，可以准确地满足用户对环境的舒适性要求，提供了用户的使用体验。本实施例的方法还可以对于使用记录比较独特的用户，建立个性化的画像模型，满足这些用户的智能要求。
至此，本领域技术人员应认识到，虽然本文已详尽示出和描述了本发明的多个示例性实施例，但是，在不脱离本发明精神和范围的情况下，仍可根据本发明公开的内容直接确定或推导出符合本发明原理的许多其他变型或修改。因此，本发明的范围应被理解和认定为覆盖了所有这些其他变型或修改。

Claims (10)

  1. 一种空调器的智能控制方法，包括：
    获取目标受控空调器所在区域在设定时间段内的环境数据记录；
    利用预先训练的自调整模型对所述环境数据记录进行预测，得到自调整策略；以及
    按照所述自调整策略对所述目标受控空调器进行控制。
  2. 根据权利要求1所述的方法，其中，所述获取所述目标受控空调器所在区域在设定时间段内的环境数据记录的步骤包括：
    获取所述目标受控空调器的位置信息，根据所述位置信息获取所述设定时间段内的室外环境记录；以及
    获取所述目标受控空调器上传的所述设定时间段内的室内环境记录。
  3. 根据权利要求1所述的方法，其中，
    所述自调整策略包括以下任意一项或多项：空调器的开关机时刻、空调器的开机初始模式、空调器的开机初始参数、空调器对环境检测数据变化的响应动作。
  4. 根据权利要求1所述的方法，其中，所述自调整模型的训练过程包括：
    从运行数据平台中订阅批量的空调器日志数据，所述运行数据平台用于采集并记录空调器的运行日志；
    将所述日志数据解析为结构化数据，并存储为样本数据库；
    使用所述样本数据库中的数据进行机器学习模型训练，得到所述自调整模型。
  5. 根据权利要求4所述的方法，其中，所述将所述日志数据解析为结构化数据的步骤包括：
    按照所述日志数据按照类型标签提取数据；
    将提取出的数据进行分类保存，得到备选样本数据表；
    对备选样本数据表中的数据进行统计筛选，得到所述结构化数据。
  6. 根据权利要求1所述的方法，其中，在所述按照所述自调整策略对所述空调器进行控制的步骤之后还包括：
    获取所述目标受控空调器上传的运行记录；
    从所述运行记录中提取手动调整记录；
    在所述手动调整记录超过设定次数阈值后，利用所述运行记录建立所述 目标受控空调器的个性调整模型，以利用所述个性调整模型制定所述目标受控空调器的后续控制策略。
  7. 根据权利要求1所述的方法，其中，在所述按照所述自调整策略对所述空调器进行控制的步骤之后还包括：
    获取所述目标受控空调器上传的运行记录；
    从所述运行记录中提取手动调整记录；
    在所述手动调整记录超过设定次数阈值后，利用所述目标受控空调器的用户行为记录建立所述目标受控空调器的用户画像模型，以利用所述用户画像模型制定所述目标受控空调器的后续控制策略。
  8. 根据权利要求6所述的方法，其中，利用所述目标受控空调器的用户行为记录建立所述目标受控空调器的用户画像模型的步骤包括：
    获取目标用户在设定时间段内的行为记录，所述行为记录至少包括空调器使用记录；
    将所述行为记录以及所述目标受控空调器的运行记录作为训练样本，训练得到所述画像模型。
  9. 根据权利要求8所述的方法，其中，
    所述行为记录还包括以下任意一项或多项：用户的出行记录、用户的位置记录、用户的生理特征记录。
  10. 一种空调器的智能控制设备，包括：
    处理器；以及
    存储器，所述存储器内存储有控制程序，所述控制程序被所述处理器执行时用于实现根据权利要求1至9中任一项所述的空调器的智能控制方法。
PCT/CN2021/078687 2020-03-09 2021-03-02 空调器的智能控制方法与空调器的智能控制设备 WO2021179958A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010158824.7 2020-03-09
CN202010158824.7A CN113375300A (zh) 2020-03-09 2020-03-09 空调器的智能控制方法与空调器的智能控制设备

Publications (1)

Publication Number Publication Date
WO2021179958A1 true WO2021179958A1 (zh) 2021-09-16

Family

ID=77568643

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/078687 WO2021179958A1 (zh) 2020-03-09 2021-03-02 空调器的智能控制方法与空调器的智能控制设备

Country Status (2)

Country Link
CN (1) CN113375300A (zh)
WO (1) WO2021179958A1 (zh)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114543303A (zh) * 2022-01-26 2022-05-27 深圳达实智能股份有限公司 基于运行大数据的中央空调制冷站运行优化方法和系统
CN114838470A (zh) * 2022-03-04 2022-08-02 五邑大学 一种暖通空调的控制方法和系统
CN115682207A (zh) * 2023-01-04 2023-02-03 江门市恒天科技有限公司 基于用户使用偏好的加湿器智能控制方法
WO2023087692A1 (zh) * 2021-11-16 2023-05-25 青岛海尔空调器有限总公司 用于空调器的控制方法及装置、空调器、存储介质

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114322260B (zh) * 2021-12-21 2023-09-08 上海美控智慧建筑有限公司 空调自动驾驶、模型训练和预测方法、装置及设备
CN115437302B (zh) * 2022-10-21 2023-05-09 深圳昌恩智能股份有限公司 一种大型中央空调ai智能控制方法及系统

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH04316947A (ja) * 1991-04-15 1992-11-09 Matsushita Electric Ind Co Ltd 空気調和機の制御装置
CN103398451A (zh) * 2013-07-12 2013-11-20 清华大学 基于学习用户行为的多维舒适度室内环境控制方法及系统
CN108361927A (zh) * 2018-02-08 2018-08-03 广东美的暖通设备有限公司 一种基于机器学习的空调器控制方法、装置以及空调器
CN109520071A (zh) * 2018-10-18 2019-03-26 珠海东之尼电子科技有限公司 一种基于支持向量机学习的空调自适应控制方法及系统
CN110131843A (zh) * 2019-05-15 2019-08-16 珠海格力电器股份有限公司 基于大数据的空调的智能调控方法及系统
CN110410964A (zh) * 2019-06-27 2019-11-05 青岛海尔空调器有限总公司 空调器的控制方法及控制系统
CN110687802A (zh) * 2018-07-06 2020-01-14 珠海格力电器股份有限公司 一种智能家电控制方法及智能家电控制装置
CN110736232A (zh) * 2019-10-29 2020-01-31 珠海格力电器股份有限公司 空调控制方法及装置

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107423442B (zh) * 2017-08-07 2020-09-25 火烈鸟网络(广州)股份有限公司 基于用户画像行为分析的应用推荐方法及系统，储存介质及计算机设备
CN108444052B (zh) * 2018-02-28 2019-05-07 平安科技（深圳）有限公司 一种空调的控制方法及空调

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH04316947A (ja) * 1991-04-15 1992-11-09 Matsushita Electric Ind Co Ltd 空気調和機の制御装置
CN103398451A (zh) * 2013-07-12 2013-11-20 清华大学 基于学习用户行为的多维舒适度室内环境控制方法及系统
CN108361927A (zh) * 2018-02-08 2018-08-03 广东美的暖通设备有限公司 一种基于机器学习的空调器控制方法、装置以及空调器
CN110687802A (zh) * 2018-07-06 2020-01-14 珠海格力电器股份有限公司 一种智能家电控制方法及智能家电控制装置
CN109520071A (zh) * 2018-10-18 2019-03-26 珠海东之尼电子科技有限公司 一种基于支持向量机学习的空调自适应控制方法及系统
CN110131843A (zh) * 2019-05-15 2019-08-16 珠海格力电器股份有限公司 基于大数据的空调的智能调控方法及系统
CN110410964A (zh) * 2019-06-27 2019-11-05 青岛海尔空调器有限总公司 空调器的控制方法及控制系统
CN110736232A (zh) * 2019-10-29 2020-01-31 珠海格力电器股份有限公司 空调控制方法及装置

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2023087692A1 (zh) * 2021-11-16 2023-05-25 青岛海尔空调器有限总公司 用于空调器的控制方法及装置、空调器、存储介质
CN114543303A (zh) * 2022-01-26 2022-05-27 深圳达实智能股份有限公司 基于运行大数据的中央空调制冷站运行优化方法和系统
CN114838470A (zh) * 2022-03-04 2022-08-02 五邑大学 一种暖通空调的控制方法和系统
CN115682207A (zh) * 2023-01-04 2023-02-03 江门市恒天科技有限公司 基于用户使用偏好的加湿器智能控制方法

Also Published As

Publication number Publication date
CN113375300A (zh) 2021-09-10

Similar Documents

Publication Publication Date Title
WO2021179958A1 (zh) 空调器的智能控制方法与空调器的智能控制设备
KR102553299B1 (ko) 데이터 학습 서버 및 이의 학습 모델 생성 및 이용 방법
CN112128936B (zh) 空调器的智能控制方法与智能控制设备
CN110410964B (zh) 空调器的控制方法及控制系统
CN113377018A (zh) 空调器的智能控制方法与空调器的智能控制设备
US20210190361A1 (en) Air conditioner
WO2022041987A1 (zh) 空调器的智能控制方法与智能控制设备
CN109471370B (zh) 一种基于排气扇操作数据的行为预测及控制方法、系统
CN109357352B (zh) 基于决策树分类算法的空调智能温控系统及方法
KR20220071405A (ko) 스마트 팜의 빅 데이터를 이용한 농업 지원 시스템 및 방법
CN110925936A (zh) 空调控制方法、装置、计算机设备和存储介质
CN112128935B (zh) 空调器的智能控制方法与智能控制设备
CN112198853A (zh) 智能家居设备的控制方法和装置
CN115905688B (zh) 一种基于人工智能和智能家居的参考信息生成方法
CN116520718A (zh) 一种基于物联网技术的智能家居自适应控制方法和装置
Yin et al. Personalized ambience: an integration of learning model and intelligent lighting control
KR102145777B1 (ko) 인공지능을 이용한 셀프검증형 감성분석에 기반한 기기제어 방법 및 장치
CN110598916A (zh) 一种用户行为模型的构建方法及系统
US20230301247A1 (en) Machine learning systems for autonomous and semiautonomous plant growth
CN114281192B (zh) 家居设备控制方法、装置、电子设备及存储介质
RU2772237C2 (ru) Сервер обучения на основе данных и способ формирования и использования его обучающей модели
CN117762032B (zh) 基于场景适应和人工智能的智能设备控制系统及方法
CN118011844A (zh) 调节方法、装置、设备及存储介质
CN116659069A (zh) 一种空调控制方法、装置、存储介质及空调
CN116130102A (zh) 睡眠环境数据的确定方法及装置、存储介质及电子装置

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21768054

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 21768054

Country of ref document: EP

Kind code of ref document: A1