WO2019196488A1 - 一种控制家用电器执行控制指令的方法及装置 - Google Patents
一种控制家用电器执行控制指令的方法及装置 Download PDFInfo
- Publication number
- WO2019196488A1 WO2019196488A1 PCT/CN2018/121531 CN2018121531W WO2019196488A1 WO 2019196488 A1 WO2019196488 A1 WO 2019196488A1 CN 2018121531 W CN2018121531 W CN 2018121531W WO 2019196488 A1 WO2019196488 A1 WO 2019196488A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- control instruction
- user
- home appliance
- historical behavior
- neural network
- Prior art date
Links
- 238000000034 method Methods 0.000 title claims abstract description 43
- 238000013528 artificial neural network Methods 0.000 claims abstract description 98
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 94
- 238000012360 testing method Methods 0.000 claims description 35
- 238000012549 training Methods 0.000 claims description 35
- 238000012790 confirmation Methods 0.000 claims description 12
- 238000012545 processing Methods 0.000 claims description 10
- 230000006399 behavior Effects 0.000 description 90
- 230000006870 function Effects 0.000 description 9
- 238000010586 diagram Methods 0.000 description 5
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 5
- 238000010438 heat treatment Methods 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000013473 artificial intelligence Methods 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 240000007594 Oryza sativa Species 0.000 description 1
- 235000007164 Oryza sativa Nutrition 0.000 description 1
- 230000004913 activation Effects 0.000 description 1
- 238000004378 air conditioning Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000009795 derivation Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000006698 induction Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 210000005036 nerve Anatomy 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 235000009566 rice Nutrition 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B15/00—Systems controlled by a computer
- G05B15/02—Systems controlled by a computer electric
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B19/00—Programme-control systems
- G05B19/02—Programme-control systems electric
- G05B19/04—Programme control other than numerical control, i.e. in sequence controllers or logic controllers
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B19/00—Programme-control systems
- G05B19/02—Programme-control systems electric
- G05B19/418—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS] or computer integrated manufacturing [CIM]
Definitions
- the present invention relates to the field of human-computer interaction, and in particular, to a method and apparatus for controlling a home appliance to execute a control command.
- the home appliance control APP can support various usage scenarios in which the user controls the home appliance.
- the user manually sets a control command for controlling the home appliance to perform related operations on the terminal on which the home appliance control APP is installed, and a time point at which the control instruction is executed, and then installs the home appliance control class.
- the terminal of the APP reports the control command set by the user and the time when the control command is executed to the server, and finally the server sends a control command to the home appliance according to the preset time point of the user.
- An aspect of the present invention provides a method for controlling a home appliance to execute a control instruction, including:
- the historical behavior including operation information set by the user for controlling the home appliance to execute the control instruction, and using the preset artificial neural network algorithm model to set the time for the user
- the historical behavior is analyzed to obtain a control instruction for controlling the home appliance and a time point at which the control instruction is executed; the control instruction is sent to the home appliance, and the household appliance is controlled to perform the Control instruction.
- control information set by the user during the set time period and the operation information of the time point of execution of the control instruction are acquired, and the historical behavior of the user within the set time is performed by using a preset artificial neural network algorithm model.
- the analysis realizes the automatic recording and analysis of the user's behavior habits, and obtains the control instructions that match the historical behaviors, without the need for the user to manually set up, and achieves the purpose of automatically helping the user to set the operation information, thereby improving the user experience.
- the method further includes: sending the control instruction to the home appliance, and before the controlling the home appliance executes the control instruction at the time point, the method further includes:
- the prompt information is sent to the user, and the content of the prompt information includes a control instruction executed by the home appliance and a time point at which the control instruction is executed.
- the prompt information may be sent to the user, and the content of the prompt information includes the control point of the execution of the home appliance and the time when the control instruction is executed, so as to achieve the effect of reminding the user, Further enhance the user experience.
- the method further includes: sending the control instruction to the home appliance, and before the controlling the home appliance executes the control instruction at the time point, the method further includes:
- the confirmation instruction is for indicating that the home appliance is allowed to execute the control instruction at the time point.
- the preset artificial neural network algorithm model is used to analyze the historical behavior of the user within a set time, and obtain a control instruction that matches the historical behavior and a time point at which the control instruction is executed, including:
- An initial artificial neural network algorithm model is established according to the historical behavior of the user within a set time period, and the input of the initial artificial neural network algorithm model is a historical behavior of the user, and the output is a control instruction matching the historical behavior and the a time point at which the control instruction is executed; a historical instruction of the user, a control instruction matching the historical behavior, and a time point at which the control instruction is executed as a candidate sample; selecting a training sample in the candidate sample, Performing an initial artificial neural network algorithm model for training, selecting a test sample among the candidate samples other than the training sample, and using the test sample to perform positive on the initial artificial neural network algorithm model Testing the network to obtain an artificial neural network algorithm model; using the artificial neural network algorithm model, determining a control instruction that matches the historical behavior and a time point at which the control instruction is executed.
- the present invention by inputting the historical behavior of the user in the set time period as an input parameter into the initial model of the artificial neural network algorithm, artificial intelligence analysis is performed, and the control instruction matching the historical behavior and the execution of the control instruction are obtained.
- the time point thus achieving automatic recording and analysis of the user's historical behavior.
- the forward network test finally obtains an artificial neural network algorithm model, and uses the artificial neural network algorithm model to determine a control instruction matching the historical behavior and a time point at which the control instruction is executed, thereby realizing automatic help of the user to set and control the household appliance.
- the instruction to perform the operation does not require the user to manually set it, which improves the user experience.
- the method further includes:
- the artificial neural network algorithm model is modified using a control command executed by the home appliance at the point in time.
- the artificial neural network algorithm model is modified by using the control instruction executed by the household appliance at the time of executing the control instruction, and the accuracy of the artificial network algorithm model can be improved.
- the operation information set by the user for controlling the home appliance to execute the control instruction to the home appliance includes at least a time point of the home appliance executing the control instruction, a control instruction executed by the home appliance, and at least a position information of the home appliance executing the control instruction.
- Another aspect of the present invention provides an apparatus for controlling a home appliance to execute a control instruction, including:
- An obtaining unit configured to acquire a historical behavior of the user within a set time period, where the historical behavior includes operation information set by the user for the home appliance for controlling the home appliance to execute the control instruction;
- a processing unit configured to analyze, by using a preset artificial neural network algorithm model, the historical behavior of the user within a set time, to obtain a control instruction that matches the historical behavior, and a time point at which the control instruction is executed; And sending the control instruction to the home appliance; and the control unit is configured to control the home appliance to execute the control instruction at the time point.
- the sending unit is further configured to: send prompt information to the user, where the content of the prompt information includes a control instruction executed by the home appliance and a time point at which the control instruction is executed.
- the device further includes: a determining unit, configured to determine that the confirmation instruction of the prompt information is received by the user, where the confirmation instruction is used to indicate that the home appliance is allowed to execute the control instruction at the time point .
- a determining unit configured to determine that the confirmation instruction of the prompt information is received by the user, where the confirmation instruction is used to indicate that the home appliance is allowed to execute the control instruction at the time point .
- the processing unit is configured to analyze, by using a preset artificial neural network algorithm model, the historical behavior of the user within a set time, and obtain a control instruction matching the historical behavior and the control.
- an initial artificial neural network algorithm model according to the historical behavior of the user within a set time period, where the input of the initial artificial neural network algorithm model is a historical behavior of the user, and the output is a control instruction matching the historical behavior. And a time point at which the control instruction is executed; a history instruction of the user, an operation indicated by the operation information in the historical behavior of the user, a control instruction for controlling the operation of the home appliance as a candidate sample; and selecting a training sample in the candidate sample And training the initial artificial neural network algorithm model, and selecting a test sample among the candidate samples other than the training sample, and using the test sample, the initial artificial neural network algorithm
- the model performs a forward network test to obtain an artificial neural network algorithm model; using the artificial neural network algorithm model, a control instruction matching the historical behavior and a time point at which the control instruction is executed are determined.
- the processing unit is further configured to: modify the artificial neural network algorithm model by using a control instruction executed by the home appliance at the time point.
- control instruction set by the user on the home appliance and the operation information at the time point when the control instruction is executed include a time point at which the home appliance executes the control instruction, a control instruction executed by the home appliance, and a location information of the home appliance executing the control instruction. At least one of them.
- the present invention provides a method and apparatus for controlling a home appliance to execute a control command, by acquiring operation information set by a user for controlling a home appliance to execute a control command within a set time period, and using a preset artificial nerve
- the network algorithm model analyzes the historical behavior of the user during the set time, obtains a control instruction that matches the historical behavior, and a time point at which the control instruction is executed, and sends a control instruction to the home appliance to control the execution of the control device by the home appliance.
- control instruction is executed, thereby realizing the automatic recording and analysis of the user's behavior habits, which can automatically help the user to set the time point and operation instruction of the home appliance to perform the operation, and the user is not required to manually operate, thereby bringing convenience to the user, thereby improving The user experience.
- FIG. 1 is a flowchart of a method for controlling a home appliance to execute a control instruction according to an embodiment of the present invention
- FIG. 2 is a schematic diagram of a historical artificial neural network algorithm model for analyzing a user's historical behavior during a set time according to an embodiment of the present invention, and obtaining a control instruction that matches a historical behavior and a time point at which the control instruction is executed.
- FIG. 3 is a structural block diagram of an apparatus for controlling a home appliance to execute a control instruction according to an embodiment of the present invention.
- the home appliance control APP on the application market will have a scene selection mode, such as: going home, leaving home mode, or providing a user-editable scene mode.
- scene modes are based on the user manually setting when the home appliance is turned on and when to turn off, and the time point of the manual reservation setting is sent to the server, and the server uniformly issues the control command according to the preset time point of the user.
- the control command set by the user is executed at the time point set by the user.
- This implementation requires the user to manually add the home appliance and requires the user to manually set the time. This method is mechanical, and in the case of a large number of home appliances, the manual addition method is cumbersome.
- an embodiment of the present invention provides a method and apparatus for controlling a home appliance to execute a control instruction, and by automatically recording and analyzing a historical behavior of a user within a set time period, obtaining a control instruction that matches a historical behavior and And controlling a time point of execution of the instruction, and sending a control instruction to the home appliance, and controlling the home appliance to execute the control instruction at a time point, thereby realizing an operation information for automatically assisting the user to set the home appliance to execute the control instruction, and controlling the automatic operation of the home appliance, Improved user experience without the need for manual setup by the user.
- FIG. 1 is a schematic diagram of a method for controlling a home appliance to execute a control instruction according to an embodiment of the present invention.
- the execution body of the method shown in FIG. 1 may be a control device. Referring to FIG. 1, the method includes:
- S101 Acquire a historical behavior of the user within a set time period.
- the historical behavior of the user in the set time period may be acquired, and the historical behavior may include operation information set by the user on the home appliance for controlling the home appliance to execute the control instruction.
- the set time period may be one month, or may be half a year, one year, etc., and the behavior within the set time period needs to reflect the long-term behavior habit of the user.
- the historical behavior of the user can be understood as operation information set by the user for controlling the home appliance to execute the control instruction for the home appliance within the set time period.
- the operation information may include a time when the user-set home appliance performs an operation, an operation instruction executed, a category of the home appliance, and the like, which are not limited by the embodiment of the present invention.
- the user sets the air conditioner in the home appliance and the specific information set may include: an operation mode, a temperature, a time, and the like of the air conditioner, and the specific information of the setting may be understood as a scene mode of the air conditioner.
- the user sets the air conditioner temperature to 30 degrees Celsius at 21:00 after going home from work every night, and the mode is set to the heating mode.
- the user can also set the air conditioning off time or adjust to other scenes.
- the user when the user sets the operation information for controlling the home appliance to execute the control instruction, the user may use the APP on the terminal to set, or use the remote controller to set, or use the switch on the home appliance to set.
- the other embodiments are not limited in this embodiment of the present invention.
- S102 Analyze a historical behavior of the user in a set time by using a preset artificial neural network algorithm model, and obtain a control instruction that matches the historical behavior and a time point at which the control instruction is executed.
- the preset artificial neural network algorithm model may be used to analyze the historical behavior of the user within the set time, and the historical behavior is obtained. Matching control instructions and the point in time at which the control instructions are executed.
- the preset artificial neural network algorithm model is used to analyze the historical behavior of the user in the set time, and the control instruction matching the historical behavior and the time point of executing the control instruction are obtained. And sending a control instruction to the household appliance to control the household appliance to execute the control instruction at the time point.
- S103 Send a control instruction to the home appliance, and control the home appliance to execute the control instruction at a time point when the control instruction is executed.
- control home appliance executes the control instruction at the time point when the control instruction is executed, it is necessary to send a control instruction to the home appliance.
- the water heater needs to be turned on when the time reaches 7:00 in the morning.
- the historical behavior of the user in the set time period is obtained, and the historical behavior of the user in the set time is analyzed by using a preset artificial neural network algorithm model, and the control matching the historical behavior is obtained.
- the instruction and the time point when the control instruction is executed send a control instruction to the home appliance, and control the household appliance to execute the control instruction at the time point when the control instruction is executed, thereby realizing automatic recording and analysis of the user behavior habit, thereby automatically helping the user to set and control the household appliance.
- the operation information of the operation is performed without the user's manual operation, which brings convenience to the user, thereby improving the user experience.
- control unit sends a control instruction to the home appliance to control the home appliance to execute the control instruction at a time point when the control instruction is executed
- the method further includes: sending the prompt information to the user, where the prompt information includes the household The control command executed by the appliance and the time point at which the control command is executed.
- the control instruction is sent to the home appliance, and the control household appliance may send the prompt information to the user before executing the control instruction at the time point when the control instruction is executed.
- the content of the prompt information includes a control instruction executed by the home appliance and a time point at which the control instruction is executed, so as to achieve the effect of reminding the user, which can further improve the user experience.
- the control instruction is sent to the home appliance, and the control home appliance may determine to receive the confirmation instruction of the prompt information by the user before executing the control instruction at the time point when the control instruction is executed.
- the confirmation command is for instructing the home appliance to execute the control instruction and the control instruction at the point in time.
- the preset artificial neural network algorithm model can be used to analyze the historical behavior of the user in the set time, and the control instruction matching the historical behavior and the time point of the execution of the control instruction are obtained.
- FIG. 2 is a schematic diagram of a historical artificial neural network algorithm model for analyzing a user's historical behavior during a set time according to an embodiment of the present invention, and obtaining a control instruction and a control instruction execution that match the historical behavior.
- the method flow chart of the time point, as shown in FIG. 2, the method includes:
- S201 Establish an initial artificial neural network algorithm model according to historical behavior of the user within a set time period.
- the historical behavior of the user in the set time period can be understood as the behavior habit of the user.
- an initial artificial neural network algorithm model can be established according to the historical behavior of the user within a set time period.
- the input parameter in the initial artificial neural network algorithm model may be the historical behavior of the user.
- the output parameter may be a control instruction matching the historical behavior and a time point at which the control instruction is executed. .
- the behavior habits of the user during the set time period can be obtained.
- the initial artificial neural network algorithm model can be determined according to different time points and the number of times the user uses the home appliance.
- the parameters set by the user to control the air conditioner start-up, heating mode, mid-range wind, etc., which affect the air conditioner may determine the initial artificial neural network algorithm model according to the parameter information set by the user.
- the user can select and control the home appliance to perform an operation through an APP that controls the home appliance on the smart phone.
- the operation information set by the user on the APP is: air conditioner, heating mode, 30 degrees Celsius
- the operation information can be uploaded to the cloud server, and the cloud server can automatically record the number of operations, time, and the like set by the user.
- the cloud server can store the operation information set by the user through the account that the user logs in, and save the operation information in the database, and the operation information can be used as an input parameter of the initial artificial neural network algorithm model.
- the input parameters in the initial artificial neural network algorithm model include, but are not limited to, one or more of the following: for example, the user opens the water heater hot water in a few minutes in the morning, and opens the induction cooker in a few minutes. When the rice is opened, the room lights are turned on, and when the user goes to work, the household appliances are automatically turned off.
- the input parameter in the initial artificial neural network algorithm model is a historical behavior of the user, and the parameter may be a single parameter or a one-dimensional array or a multi-dimensional array composed according to a certain rule, which is not limited in the embodiment of the present invention.
- a historical behavior of the user, a control instruction matching the historical behavior, and a time point at which the control instruction is executed may be used as a candidate sample.
- S203 selecting a training sample in the candidate sample, training the initial artificial neural network algorithm model, selecting a test sample in the candidate sample other than the training sample, and using the test sample, the initial artificial neural network algorithm model Perform a forward network test to obtain an artificial neural network algorithm model.
- part of the candidate sample data is used as training sample data, and a part is used as test sample data.
- the training sample data may be selected in the candidate sample data, and the initial artificial neural network algorithm model is trained.
- W K weight of the network
- bl offset.
- the weight W K of the network can be updated according to the following manner, and the offset bl:
- C(w,b) is the error energy function (taking the standard variance function as an example)
- n is the total number of training samples, and summed on the total training sample x:
- test error When testing the network with test sample data, if the test error does not meet the requirements, repeat the above steps to retrain the artificial neural network; if the test error meets the requirements, the artificial neural network training test is completed.
- the training samples may be selected in the candidate samples, the initial artificial neural network algorithm model is trained, and the test samples are selected among the candidate samples except the training samples, and the test samples are used to initialize the samples.
- the artificial neural network algorithm model is used for forward network testing to obtain an artificial neural network algorithm model.
- the purpose of performing forward network testing on the initial artificial neural network algorithm model is to determine whether the trained model is correct.
- the historical behavior of the user in the set time period is input as an input parameter into the initial model of the artificial neural network algorithm, and artificial intelligence analysis is performed, and the control instruction and the control instruction execution matched with the historical behavior are obtained.
- automatic recording and analysis of the user's historical behavior is achieved.
- the training samples are selected in the historical behavior of the user, the control commands for controlling the household appliances, and the control instructions are executed, the initial artificial neural network is trained, and the test samples other than the training samples are selected for forward network testing, and finally An artificial neural network algorithm model is obtained, and the artificial neural network algorithm model is used to determine the control instruction that matches the historical behavior and the time point at which the control instruction is executed, thereby realizing an instruction for automatically assisting the user to set and control the execution of the home appliance without the user. Going to the manual setting improves the user experience.
- S204 Using an artificial neural network algorithm model, determining a control instruction that matches the historical behavior and a time point at which the control instruction is executed.
- the artificial neural network algorithm model can be used to determine the control point for controlling the home appliance and the time point at which the control instruction is executed. It can be understood that the artificial neural network algorithm model in the embodiment of the present invention utilizes the deep learning method to learn and record the behavior habits of the user, and realizes the automatic help user to set the operation information of the household appliance, without the user manually setting, thereby improving The user experience.
- the control home appliance executes the control instruction at a time point when the control instruction is executed, the method further includes: using the control instruction executed by the home appliance at the time point, to the artificial neural network The algorithm model is modified.
- the artificial neural network algorithm model is modified by using the control instruction executed by the household appliance at the time when the control instruction is executed, and the accuracy of the artificial network algorithm model can be improved.
- the operation information set by the user for controlling the home appliance to execute the control instruction for the home appliance includes at least one of a time point at which the home appliance executes the control command, a control command executed by the home appliance, and position information of the home appliance executing the control command item.
- the operation information set by the user for controlling the execution instruction of the home appliance can be input into the artificial neural network algorithm, and the intelligent control of the artificial neural network algorithm is used to obtain the control instruction matching the historical behavior. And the point in time at which the control instruction is executed.
- the artificial neural network algorithm may be directly integrated in the controller of the home appliance, or may be an independent smart device, by which the operation information set by the user for controlling the operation of the home appliance is input into the artificial neural network algorithm.
- the smart device can include: a wireless communication module, a router, a server, a smart phone, and the like.
- the household appliance air conditioner has a wireless communication module, and the operation information of the air conditioner performing the operation set by the user can be sent to the smart device, and the smart device inputs the operation information set by the user for controlling the operation of the home appliance to the artificial neural network.
- the algorithm In the algorithm.
- the APP on the terminal can obtain the control instruction for controlling the operation of the home appliance by scanning, and on the display interface of the APP.
- the function button for the user to click and operate can be displayed, and the user can select the operation information for controlling the operation of the home appliance in the function operation button.
- the embodiment of the present invention further provides a device for controlling a home appliance to execute a control command according to the same concept as the method for controlling a home appliance to execute a control command
- FIG. 3 is a control home according to an embodiment of the present invention.
- a structural block diagram of an apparatus for executing a control instruction by an electric appliance as shown in FIG. 3, the apparatus includes: an obtaining unit 101, a processing unit 102, a transmitting unit 103, and a control unit 104, wherein:
- the obtaining unit 101 is configured to acquire a historical behavior of the user within a set time period, where the historical behavior includes operation information set by the user for the home appliance for controlling the home appliance to execute the control instruction.
- the processing unit 102 is configured to analyze, by using a preset artificial neural network algorithm model, the historical behavior of the user acquired by the obtaining unit 101 within a set time period, and obtain a control instruction and a control instruction executed to match the historical behavior. Time point.
- the sending unit 103 is configured to send a control instruction to the home appliance.
- the control unit 104 is configured to control the home appliance to execute the control instruction at a time point.
- the sending unit 103 is further configured to send prompt information to the user, where the content of the prompt information includes a control instruction executed by the home appliance according to the control instruction and a time point at which the control instruction is executed.
- the device further includes: a determining unit 105, configured to determine that the confirmation instruction of the prompt information is received by the user, and the confirmation instruction is used to indicate that the home appliance is allowed to execute the control instruction at a time point when the control instruction is executed.
- a determining unit 105 configured to determine that the confirmation instruction of the prompt information is received by the user, and the confirmation instruction is used to indicate that the home appliance is allowed to execute the control instruction at a time point when the control instruction is executed.
- the processing unit 102 is configured to analyze the historical behavior of the user in the set time by using a preset artificial neural network algorithm model as follows, and obtain a control instruction that matches the historical behavior and a time point at which the control instruction is executed:
- an initial artificial neural network algorithm model is established.
- the input of the initial artificial neural network algorithm model is the historical behavior of the user, and the output is a control instruction and control that matches the historical behavior.
- the time point at which the instruction is executed; the time history of the user's historical behavior, the control instruction matching the historical behavior, and the execution of the control instruction are taken as candidate samples; the training samples are selected in the candidate samples, and the initial artificial neural network algorithm model is trained.
- test samples in the candidate samples other than the training samples using the test samples, performing a forward network test on the initial artificial neural network algorithm model to obtain an artificial neural network algorithm model; using an artificial neural network algorithm model Determining a control instruction that matches the historical behavior and a point in time at which the control instruction is executed.
- processing unit 102 is further configured to: modify the artificial neural network algorithm model by using a control instruction executed by the home appliance at the time point.
- the operation information set by the user for controlling the home appliance to execute the control instruction to the home appliance includes at least a time point at which the home appliance executes the control instruction, a control instruction executed by the home appliance, and at least a position information of the home appliance executing the control instruction.
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- General Engineering & Computer Science (AREA)
- Manufacturing & Machinery (AREA)
- Quality & Reliability (AREA)
- Selective Calling Equipment (AREA)
- Telephonic Communication Services (AREA)
Abstract
一种控制家用电器执行控制指令的方法及装置，通过获取用户在设定时间段内的历史行为，历史行为包括用户对家用电器设置的用于控制家用电器执行控制指令的操作信息（S101），并利用预设的人工神经网络算法模型对用户在设定时间内的历史行为进行分析，得到与用户历史行为相匹配的控制指令以及控制指令执行的时间点（S102），向家用电器发送控制指令，控制家用电器在控制指令执行的时间点执行控制指令（S103），从而实现自动帮助用户设置家用电器的操作信息，无需用户手动操作，进而提升用户体验。
Description
本申请要求于2018年4月13日提交中国专利局、申请号为201810332006.7、发明名称为“一种控制家用电器执行控制指令的方法及装置”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
本发明涉及人机交互领域，尤其涉及一种控制家用电器执行控制指令的方法及装置。
随着经济的发展，用户使用的家用电器的种类也越来越多。为了满足用户的需求，家用电器的功能也越来越齐全。为了方便人们的日常生活，具有对各种家用电器进行控制的家用电器控制类APP(Application，应用程序)被应用。
目前，家用电器控制类APP可支持用户对家用电器进行控制的各种使用场景。但是大部分使用场景下，都是基于用户在安装有家用电器控制类APP的终端上预先手动设置控制家用电器执行相关操作的控制指令，以及控制指令执行的时间点，之后安装有家用电器控制类APP的终端将用户设置的控制指令以及控制指令执行的时间点上报给服务器，最后由服务器按照用户预先设定的时间点统一向家用电器下发控制指令。
但是，由于用户需要在家用电器控制类APP上手动添加需要使用的各种家用电器，并在每次使用时提前为每种家用电器手动设置执行相关操作的控制指令，以及控制指令执行的时间点等，若家用电器的种类比较多，则会导致操作过程过于繁琐，这种手动设置的实现方式比较机械，用户体验较差。
发明内容
本发明的目的是提供一种控制家用电器执行控制指令的方法及装置，以实现对用户习惯的自动记录与分析，提升用户体验。
本发明的目的是通过以下技术方案实现的：
本发明一方面提供一种控制家用电器执行控制指令的方法，包括：
获取用户在设定时间段内的历史行为，所述历史行为包括用户对家用电器设置的用于控制家用电器执行控制指令的操作信息；利用预设的人工神经网络算法模型对用户在设定时间内的所述历史行为进行分析，得到用于控制家用电器的控制指令以及所述控制指令执行的时间点；向家用电器发送所述控制指令，控制所述家用电器在所述时间点执行所述控制指令。
本发明中通过获取用户在设定时间段内设置的控制指令以及所述控制指令执行的时间点的操作信息，并利用预设的人工神经网络算法模型对用户在设定时间内的历史行为进行分析，从而实现了对用户行为习惯的自动记录与分析，得到与历史行为相匹配的控制指令，无需用户手动设置，达到自动帮助用户设置操作信息的目的，提升了用户体验。
可选的，向家用电器发送所述控制指令，控制家用电器在所述时间点执行所述控制指令之前，所述方法还包括：
向用户发送提示信息，所述提示信息提示的内容包括家用电器执行的控制指令以及所述控制指令执行的时间点。
本发明中控制家用电器依据控制指令执行操作之前，可向用户发送提示信息，提示信息提示的内容包括家用电器在执行的控制指令以及控制指令执行的时间点，达到对用户进行提醒的效果，可进一步提升用户体验。
可选的，向家用电器发送所述控制指令，控制家用电器在所述时间点执行所述控制指令之前，所述方法还包括：
确定接收到用户对所述提示信息的确认指令，所述确认指令用于指示允许所述家用电器在所述时间点执行所述控制指令。
可选的，利用预设的人工神经网络算法模型对用户在设定时间内的所述历史行为进行分析，得到与历史行为相匹配的控制指令以及所述控制指令执行的时间点，包括：
根据用户在设定时间段内的所述历史行为，建立初始人工神经网络算法模型，所述初始人工神经网络算法模型的输入为用户的历史行为，输出为与历史行为相匹配的控制指令以及所述控制指令执行的时间点；将用户的历史行为、与所述历史行为相匹配的控制指令以及所述控制指令执行的时间点，作为候选样本；在所述候选样本中选择训练样本，对所述初始人工神经网络算法模型进行训练，并在所述候选样本中除所述训练样本之外的其它样本中选择测试样本，并利用所述测试样本，对所述初始人工神经网络算法模型进行正向网络测试，得到人工神经网络算法模型；利用所述人工神经网络算法模型，确定与历史行为相匹配的控制指令以及所述控制指令执行的时间点。
本发明中通过将用户在设定时间段内的历史行为作为输入参数输入到人工神经网络算法初始模型中，进行人工智能分析，输出得到与历史行为相匹配的控制指令以及所述控制指令执行的时间点，从而实现了对用户历史行为的自动记录与分析。其次，在用户的历史行为、与所述历史行为相匹配的控制指令以及所述控制指令执行的时间点中选择训练样本，对初始人工神经网络进行训练，并选择除训练样本外的测试样本进行正向网络测试，最终得到人工神经网络算法模型，利用该人工神经网络算法模型，确定与历史行为相匹配的控制指令以及所述控制指令执行的时间点，从而实现了自动帮助用户设置控制家用电器执行操作的指令，无需用户再去手动设置，提升了用户体验。
可选的，向家用电器发送所述控制指令，控制家用电器在所述时间点执行所述控制指令之后，所述方法还包括：
利用家用电器在所述时间点执行的控制指令，对所述人工神经网络算法模型进行修正。
本发明中利用家用电器在执行控制指令的时间点执行的控制指令，对人工 神经网络算法模型进行修正，可提高人工网络算法模型的准确度。
可选的，用户对家用电器设置的用于控制家用电器执行控制指令的操作信息包括家用电器执行控制指令的时间点、家用电器执行的控制指令、执行控制指令的家用电器的位置信息中的至少一项。
本发明另一方面提供一种控制家用电器执行控制指令的装置，包括：
获取单元，用于获取用户在设定时间段内的历史行为，所述历史行为包括用户对家用电器设置的用于控制家用电器执行控制指令的操作信息；
处理单元，用于利用预设的人工神经网络算法模型对用户在设定时间内的所述历史行为进行分析，得到与历史行为相匹配的控制指令以及所述控制指令执行的时间点；发送单元，用于向家用电器发送所述控制指令；控制单元，用于控制家用电器在所述时间点执行所述控制指令。
可选的，所述发送单元还用于：向用户发送提示信息，所述提示信息提示的内容包括家用电器执行的控制指令以及所述控制指令执行的时间点。
可选的，所述装置还包括：确定单元，用于确定接收到用户对所述提示信息的确认指令，所述确认指令用于指示允许所述家用电器在所述时间点执行所述控制指令。
可选的，所述处理单元用于按如下方式利用预设的人工神经网络算法模型对用户在设定时间内的所述历史行为进行分析，得到与历史行为相匹配的控制指令以及所述控制指令执行的时间点：
根据用户在设定时间段内的所述历史行为，建立初始人工神经网络算法模型，所述初始人工神经网络算法模型的输入为用户的历史行为，输出为与所述历史行为相匹配的控制指令以及所述控制指令执行的时间点；将用户的历史行为、用户的历史行为中操作信息所指示的操作、控制家用电器执行操作的控制指令，作为候选样本；在所述候选样本中选择训练样本，对所述初始人工神经网络算法模型进行训练，并在所述候选样本中除所述训练样本之外的其它样本中选择测试样本，并利用所述测试样本，对所述初始人工神经网络算法模型进 行正向网络测试，得到人工神经网络算法模型；利用所述人工神经网络算法模型，确定与历史行为相匹配的控制指令以及所述控制指令执行的时间点。
可选的，所述处理单元还用于：利用家用电器在所述时间点执行的控制指令，对所述人工神经网络算法模型进行修正。
可选的，用户对家用电器设置的控制指令以及所述控制指令执行的时间点的操作信息包括家用电器执行控制指令的时间点、家用电器执行的控制指令、执行控制指令的家用电器的位置信息中的至少一项。
本发明提供了一种控制家用电器执行控制指令的方法及装置，通过获取用户在设定时间段内对家用电器设置的用于控制家用电器执行控制指令的操作信息，并利用预设的人工神经网络算法模型对用户在设定时间内的历史行为进行分析，得到与历史行为相匹配的控制指令以及所述控制指令执行的时间点，向家用电器发送控制指令，控制家用电器在控制指令执行的时间点执行控制指令，从而实现了对用户行为习惯的自动记录与分析，可自动帮助用户设置家用电器执行操作的时间点、操作指令等，无需用户手动操作，为用户带来了方便，进而提升了用户体验。
图1为本发明实施例提供的一种控制家用电器执行控制指令的方法流程图；
图2为本发明实施例提供的一种利用预设的人工神经网络算法模型对用户在设定时间内的历史行为进行分析，得到与历史行为相匹配的控制指令以及控制指令执行的时间点的方法流程图；
图3为本发明实施例提供的一种控制家用电器执行控制指令的装置的结构框图。
下面将结合本发明实施例中的附图，对本发明实施例中的技术方案进行清 楚、完整地描述，显然，所描述的实施例仅仅是本发明一部分实施例，并不是全部的实施例。基于本发明中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本发明保护的范围。
目前，应用市场上的家用电器控制类APP都会带有场景选择模式，例如：回家、离家模式，或者提供用户可编辑的场景模式。但是很多场景模式都是基于用户手动设置预约家用电器在何时开启、何时关闭，并将手动预约设置的时间点发送给服务器，由服务器按照用户预先设定的时间点统一下发控制指令，在用户设置的时间点执行用户所设置的控制指令。这种实现方式需要用户手动添加家用电器并且需要用户手动设置时间，这种方式比较机械，并且在家用电器比较多的情况下，手动添加的方式比较繁琐。
有鉴于此，本发明实施例提供了一种控制家用电器执行控制指令的方法及装置，通过在设定时间段内对用户历史行为的自动记录与分析，得到与历史行为相匹配的控制指令以及所述控制指令执行的时间点，并向家用电器发送控制指令，控制家用电器在时间点执行控制指令，从而实现了自动帮助用户设置家用电器执行控制指令的操作信息，控制家用电器自动执行操作，无需用户手动设置，提升了用户体验。
图1所示为本发明实施例提供的一种控制家用电器执行控制指令的方法，图1所示方法的执行主体可以为一种控制装置，参阅图1所示，该方法包括：
S101：获取用户在设定时间段内的历史行为。
本发明实施例中，可获取用户在设定时间段内的历史行为，该历史行为可包括用户对家用电器设置的用于控制家用电器执行控制指令的操作信息。
具体的，设定时间段可以为一个月，也可以为半年、一年等，设定时间段内的行为需要能够反映出用户长期的行为习惯。用户的历史行为可理解为用户在设定时间段内对家用电器设置的用于控制家用电器执行控制指令的操作信息。
可以理解的是，该操作信息可包括用户设置的家用电器执行操作的时间、 所执行的操作指令、家用电器的类别等，本发明实施例对此不作限定。例如，用户对家用电器中的空调进行设置，所设置的具体信息可包括：空调的运行模式、温度、时间等，可将该设置的具体信息理解是空调的一种场景模式。
例如，用户在每天晚上下班回家后的21:00将空调温度设置为30摄氏度、模式设置为制热模式。当然，用户也可设置空调关闭时间或者调成其它场景。
具体的，用户对家用电器设置用于控制家用电器执行控制指令的操作信息时，可利用终端上的APP进行设置，也可利用遥控器进行设置，也可利用家用电器上的开关进行设置，也可采用其它的方式，本发明实施例对此不作限定。
S102：利用预设的人工神经网络算法模型对用户在设定时间内的历史行为进行分析，得到与历史行为相匹配的控制指令以及控制指令执行的时间点。
具体的，在获取到用户在设定时间段内对家用电器设置的操作信息之后，可利用预设的人工神经网络算法模型对用户在设定时间内的历史行为进行分析，得到与历史行为相匹配的控制指令以及控制指令执行的时间点。在本发明实施例中，可以理解为，利用预设的人工神经网络算法模型对用户在设定时间内的历史行为进行分析，得到与历史行为相匹配的控制指令以及控制指令执行的时间点，并向家用电器发送控制指令，控制家用电器在时间点执行控制指令。
S103：向家用电器发送控制指令，控制家用电器在控制指令执行的时间点执行控制指令。
具体的，在控制家用电器在控制指令执行的时间点执行控制指令时，需要向家用电器发送控制指令。
例如，用户每天在早上7:00起床需要开启热水器烧热水洗漱，那么需要在时间达到早上7:00时，控制热水器处于开启状态。
本发明实施例中，通过获取用户在设定时间段内的历史行为，并利用预设的人工神经网络算法模型对用户在设定时间内的历史行为进行分析，得到与历史行为相匹配的控制指令以及控制指令执行的时间点，向家用电器发送控制指令，控制家用电器在控制指令执行的时间点执行控制指令，从而实现对用户行 为习惯的自动记录与分析，达到自动帮助用户设置控制家用电器执行操作的操作信息，无需用户手动操作，为用户带来了方便，进而提升了用户体验。
一种可能的实施方式中，向家用电器发送控制指令，控制家用电器在控制指令执行的时间点执行控制指令之前，所述方法还包括：向用户发送提示信息，该提示信息提示的内容包括家用电器执行的控制指令以及控制指令执行的时间点。
本发明实施例中，向家用电器发送控制指令，控制家用电器在控制指令执行的时间点执行控制指令之前，可向用户发送提示信息。该提示信息提示的内容包括家用电器执行的控制指令以及所述控制指令执行的时间点，达到对用户进行提醒的效果，可进一步提升用户体验。
为了使得家用电器能够依据控制指令执行操作，向家用电器发送控制指令，控制家用电器在控制指令执行的时间点执行所述控制指令之前，可确定接收用户对所述提示信息的确认指令。该确认指令用于指示允许家用电器在所述时间点执行控制指令以及所述控制指令。
一种可能的实施方式中，可通过如下方式利用预设的人工神经网络算法模型对用户在设定时间内的历史行为进行分析，得到与历史行为相匹配的控制指令以及控制指令执行的时间点，如图2所示为本发明实施例提供的一种利用预设的人工神经网络算法模型对用户在设定时间内的历史行为进行分析，得到与历史行为相匹配的控制指令以及控制指令执行的时间点的方法流程图，参阅图2所示，该方法包括：
S201：根据用户在设定时间段内的历史行为，建立初始人工神经网络算法模型。
可以理解的是，本发明实施例中，用户在设定时间段内的历史行为可理解为用户的行为习惯。
具体的，可根据用户在设定时间段内的历史行为，建立初始人工神经网络算法模型。其中，初始人工神经网络算法模型中的输入参数可以为用户的历史 行为，经过初始人工神经网络算法模型进行智能分析之后，输出参数可以为与历史行为相匹配的控制指令以及控制指令执行的时间点。通过对用户历史行为的记录与分析，可得到用户在设定时间段内的行为习惯。
例如，可根据不同的时间点、用户对家用电器使用的次数，确定初始人工神经网络算法模型。例如，用户设置的在几时几分的控制空调开机、制热模式、中档风等对空调有影响的参数，可根据用户设置的这些参数信息确定初始人工神经网络算法模型。
一种可能的实施方式中，用户可通过智能手机上的控制家用电器的APP，选择并控制家用电器执行操作。例如，若用户在APP上设置的操作信息为：空调、制热模式、30摄氏度，则该操作信息可被上传到云服务器，云服务器可自动记录用户设置的操作次数、时间等。
可以理解的是，云服务器可通过用户登录的账号存储用户所设置的操作信息，并将操作信息保存在数据库中，该操作信息可作为初始人工神经网络算法模型的输入参数。
具体的，本发明实施例中，初始人工神经网络算法模型中的输入参数包括但不限于如下的一种或者多种：例如，用户早上起来几时几分打开热水器热水、几时几分打开电磁炉煮饭、几时几分打开房间灯光、几时几分用户上班之后家用电器自动关闭等。初始人工神经网络算法模型中的输入参数为用户的历史行为，该参数可以是单一参数，也可以是按照一定规律组成的一维数组或者多维数组，本发明实施例对此不作限定。
S202：将用户的历史行为、与历史行为相匹配的控制指令以及控制指令执行的时间点，作为候选样本。
具体的，本发明实施例中，可将用户的历史行为、与历史行为相匹配的控制指令以及控制指令执行的时间点，作为候选样本。
S203：在候选样本中选择训练样本，对初始人工神经网络算法模型进行训练，并在候选样本中除训练样本之外的其它样本中选择测试样本，并利用测试 样本，对初始人工神经网络算法模型进行正向网络测试，得到人工神经网络算法模型。
具体的，将候选样本数据中，其中一部分用作训练样本数据，一部分用作测试样本数据。本发明实施例中，可在候选样本数据中选择训练样本数据，对初始人工神经网络算法模型进行训练。
对训练样本进行训练的方法有多种，以下选择其中的一种训练方法进行具体说明。
输入参数x，根据激活函数、初始化的权值及偏置计算出人工神经网络的实际输出a(x)，即a(x)＝1/(1+e-z),其中Z＝W
K*x+bl。
W
K：网络的权值，bl：偏置。
判断网络的期望输出y(x)与实际输出a(x)是否满足输出精度要求即：
[根据细则20.5改正09.01.2019]
，
为目标最小误差。
[根据细则20.5改正09.01.2019]
如果满足精度要求则结束训练，如不满足则更新网络的权值W K,偏置bl，直至网络的输出误差小于为止。
如果满足精度要求则结束训练，如不满足则更新网络的权值W K,偏置bl，直至网络的输出误差小于
具体的，可根据以下方式更新网络的权值W
K,偏置bl：
C(w,b)为误差能量函数(以标准方差函数为例)，n为训练样本的总数量,在总的训练样本x上进行求和得到：
更新各层权值：
更新各层偏置：
其中：
[根据细则20.5改正09.01.2019]
Wk为初始权值，
为误差能量函数对权值的偏导数； bl为初始偏置，
为误差能量函数对偏置的偏导数。
Wk为初始权值，
[根据细则20.5改正09.01.2019]
其中，、的值可通过链式求导法则获得。
其中，
具体的，初始人工神经网络模型训练完成后，并在候选样本中除训练样本之外的其它样本中选择测试样本，并利用测试样本，对初始人工神经网络算法模型进行正向网络测试，得到人工神经网络算法模型。
在用测试样本数据正向测试网络时，若测试误差不满足要求，则重复以上步骤，重新训练人工神经网络；若测试误差满足要求，则人工神经网络训练测试完成。
本发明实施例中，可在候选样本中选择训练样本，对初始人工神经网络算法模型进行训练，并在候选样本中除训练样本之外的其它样本中选择测试样本，并利用测试样本，对初始人工神经网络算法模型进行正向网络测试，得到人工神经网络算法模型。
可以理解的是，对初始人工神经网络算法模型进行正向网络测试的目的是为了确定训练出来的模型是否正确。
本发明实施例中，通过将用户在设定时间段内的历史行为作为输入参数输入到人工神经网络算法初始模型中，进行人工智能分析，输出得到与历史行为相匹配的控制指令以及控制指令执行的时间点，从而实现了对用户历史行为的自动记录与分析。其次，在用户的历史行为、控制家用电器的控制指令以及控制指令执行的时间点中选择训练样本，对初始人工神经网络进行训练，并选择除训练样本外的测试样本进行正向网络测试，最终得到人工神经网络算法模型，利用该人工神经网络算法模型，确定与历史行为相匹配的控制指令以及控制指令执行的时间点，从而实现了自动帮助用户设置控制家用电器执行操作的指令，无需用户再去手动设置，提升了用户体验。
S204：利用人工神经网络算法模型，确定与历史行为相匹配的控制指令以及控制指令执行的时间点。
具体的，在得到人工神经网络算模型之后，可利用人工神经网络算法模型，确定用于控制家用电器的控制指令以及控制指令执行的时间点。可以理解的是，本发明实施例中人工神经网络算法模型是利用深度学习的方法对用户的行为习惯进行学习并记录分析，实现自动帮助用户设置家用电器的操作信息，无需用户手动设置，进而提升了用户体验。
进一步的，向家用电器发送控制指令，控制家用电器在控制指令执行的时间点执行控制指令之后，所述方法还包括：利用家用电器在所述时间点执行的控制指令，对所述人工神经网络算法模型进行修正。
本发明实施例中，利用家用电器在控制指令执行的时间点执行的控制指令，对人工神经网络算法模型进行修正，可提高人工网络算法模型的准确度。
进一步的，用户对家用电器设置的用于控制家用电器执行控制指令的操作信息包括家用电器执行控制指令的时间点、家用电器执行的控制指令、执行控制指令的家用电器的位置信息中的至少一项。
本发明实施例中，可将用户设置的用于控制家用电器执行控制指令的操作信息输入到人工神经网络算法中，经过人工神经网络算法的智能分析，得到与所述历史行为相匹配的控制指令以及控制指令执行的时间点。该人工神经网络算法可以直接集成在家用电器的控制器中，也可以是独立的智能装置，通过该智能装置将用户设置的用于控制家用电器执行操作的操作信息输入到人工神经网络算法中。可以理解的是，该智能装置可包括：无线通讯模块、路由器、服务器、智能手机等。
例如，家用电器空调中带有无线通讯模块，可将用户设置的空调执行操作的操作信息发送给智能装置，智能装置再将用户设置的用于控制家用电器执行操作的操作信息输入到人工神经网络算法中。
具体的，经过人工神经网络算法的智能分析，得到用于控制家用电器执行操作的控制指令之后，终端上的APP可通过扫描得到该控制家用电器执行操作的控制指令，并且在APP的显示界面上可显示有供用户点击操作的功能按 键，用户可在功能操作按键中选择控制家用电器执行操作的操作信息。
基于与上述控制家用电器执行控制指令的方法实施例相同的构思，本发明实施例还提供了一种控制家用电器执行控制指令的装置，图3所示为本发明实施例提供的一种控制家用电器执行控制指令的装置的结构框图，参阅图3所示，该装置包括：获取单元101、处理单元102、发送单元103、控制单元104，其中：
获取单元101，用于获取用户在设定时间段内的历史行为，所述历史行为包括用户对家用电器设置的用于控制家用电器执行控制指令的操作信息。
处理单元102，用于利用预设的人工神经网络算法模型对获取单元101获取到的用户在设定时间段内的历史行为进行分析，得到与所述历史行为相匹配的控制指令以及控制指令执行的时间点。
发送单元103，用于向家用电器发送控制指令。
控制单元104，用于控制家用电器在时间点执行所述控制指令。
可选的，所述发送单元103还用于向用户发送提示信息，该提示信息提示的内容包括家用电器依据控制指令执行的控制指令以及控制指令执行的时间点。
进一步的，所述装置还包括：确定单元105，用于确定接收到用户对所述提示信息的确认指令，该确认指令用于指示允许家用电器在控制指令执行的时间点执行所述控制指令。
具体的，处理单元102用于按如下方式利用预设的人工神经网络算法模型对用户在设定时间内的历史行为进行分析，得到与历史行为相匹配的控制指令以及控制指令执行的时间点：
根据用户在设定时间段内的所述历史行为，建立初始人工神经网络算法模型，初始人工神经网络算法模型的输入为用户的历史行为，输出为与所述历史行为相匹配的控制指令以及控制指令执行的时间点；将用户的历史行为、与历史行为相匹配的控制指令以及控制指令执行的时间点，作为候选样本；在候选 样本中选择训练样本，对初始人工神经网络算法模型进行训练，并在候选样本中除所述训练样本之外的其它样本中选择测试样本，利用测试样本，对初始人工神经网络算法模型进行正向网络测试，得到人工神经网络算法模型；利用人工神经网络算法模型，确定与所述历史行为相匹配的控制指令以及控制指令执行的时间点。
进一步的，处理单元102还用于：利用家用电器在所述时间点执行的控制指令，对人工神经网络算法模型进行修正。
更进一步的，用户对家用电器设置的用于控制家用电器执行控制指令的操作信息包括家用电器执行控制指令的时间点、家用电器执行的控制指令、执行控制指令的家用电器的位置信息中的至少一项。
需要说明的是，本发明实施例中上述涉及的控制家用电器执行控制指令的装置中各个单元的功能实现可以进一步参照相关方法实施例的描述，在此不再赘述。
显然，本领域的技术人员可以对本发明进行各种改动和变型而不脱离本发明的精神和范围。这样，倘若本发明的这些修改和变型属于本发明权利要求及其等同技术的范围之内，则本发明也意图包含这些改动和变型在内。
Claims (12)
- 一种控制家用电器执行控制指令的方法，其特征在于，包括：获取用户在设定时间段内的历史行为，所述历史行为包括用户对家用电器设置的用于控制家用电器执行控制指令的操作信息；利用预设的人工神经网络算法模型对用户在设定时间内的所述历史行为进行分析，得到与所述历史行为相匹配的控制指令，以及所述控制指令执行的时间点；向家用电器发送所述控制指令，控制家用电器在所述时间点执行所述控制指令。
- 如权利要求1所述的方法，其特征在于，向家用电器发送所述控制指令，控制家用电器在所述时间点执行所述控制指令之前，所述方法还包括：向用户发送提示信息，所述提示信息提示的内容包括家用电器执行的控制指令以及所述控制指令执行的时间点。
- 如权利要求2所述的方法，其特征在于，向家用电器发送所述控制指令，控制家用电器在所述时间点执行所述控制指令之前，所述方法还包括：确定接收到用户对所述提示信息的确认指令，所述确认指令用于指示允许所述家用电器在所述时间点执行所述控制指令。
- 如权利要求1所述的方法，其特征在于，利用预设的人工神经网络算法模型对用户在设定时间内的所述历史行为进行分析，得到与所述历史行为相匹配的控制指令，以及所述控制指令执行的时间点，包括：根据用户在设定时间段内的所述历史行为，建立初始人工神经网络算法模型，所述初始人工神经网络算法模型的输入为用户的历史行为，输出为与所述历史行为相匹配的控制指令以及所述控制指令执行的时间点；将用户的历史行为、与所述历史行为相匹配的控制指令以及所述控制指令执行的时间点，作为候选样本；在所述候选样本中选择训练样本，对所述初始人工神经网络算法模型进行 训练，并在所述候选样本中除所述训练样本之外的其它样本中选择测试样本，并利用所述测试样本，对所述初始人工神经网络算法模型进行正向网络测试，得到人工神经网络算法模型；利用所述人工神经网络算法模型，确定与所述历史行为相匹配的控制指令以及所述控制指令执行的时间点。
- 如权利要求4所述的方法，其特征在于，向家用电器发送所述控制指令，控制家用电器在所述时间点执行所述控制指令之后，所述方法还包括：利用家用电器在所述时间点执行的控制指令，对所述人工神经网络算法模型进行修正。
- 如权利要求1至5任一项所述的方法，其特征在于，用户对家用电器设置的用于控制家用电器执行控制指令的操作信息包括家用电器执行控制指令的时间点、家用电器执行的控制指令、执行控制指令的家用电器的位置信息中的至少一项。
- 一种控制家用电器执行控制指令的装置，其特征在于，包括：获取单元，用于获取用户在设定时间段内的历史行为，所述历史行为包括用户对家用电器设置的用于控制家用电器执行控制指令的操作信息；处理单元，用于利用预设的人工神经网络算法模型对用户在设定时间内的所述历史行为进行分析，得到与所述历史行为相匹配的控制指令，以及所述控制指令执行的时间点；发送单元，用于向家用电器发送所述控制指令；控制单元，用于控制家用电器在所述时间点执行所述控制指令。
- 如权利要求7所述的装置，其特征在于，所述发送单元还用于：向用户发送提示信息，所述提示信息提示的内容包括家用电器执行的控制指令以及所述控制指令执行的时间点。
- 如权利要求8所述的装置，其特征在于，所述装置还包括：确定单元，用于确定接收到用户对所述提示信息的确认指令，所述确认指令用于指示允许 所述家用电器在所述时间点执行所述控制指令。
- 如权利要求7所述的装置，其特征在于，所述处理单元用于按如下方式利用预设的人工神经网络算法模型对用户在设定时间内的所述历史行为进行分析，得到与所述历史行为相匹配的控制指令以及所述控制指令执行的时间点：根据用户在设定时间段内的所述历史行为，建立初始人工神经网络算法模型，所述初始人工神经网络算法模型的输入为用户的历史行为，输出为与所述历史行为相匹配的控制指令，以及所述控制指令执行的时间点；将用户的历史行为、与所述历史行为相匹配的控制指令以及所述控制指令执行的时间点，作为候选样本；在所述候选样本中选择训练样本，对所述初始人工神经网络算法模型进行训练，并在所述候选样本中除所述训练样本之外的其它样本中选择测试样本，并利用所述测试样本，对所述初始人工神经网络算法模型进行正向网络测试，得到人工神经网络算法模型；利用所述人工神经网络算法模型，确定与所述历史行为相匹配的控制指令，以及所述控制指令执行的时间点。
- 如权利要求10所述的装置，其特征在于，所述处理单元还用于：利用家用电器执行的控制指令，对所述人工神经网络算法模型进行修正。
- 如权利要求7至11任一项所述的装置，其特征在于，用户对家用电器设置的用于控制家用电器执行控制指令的操作信息包括家用电器执行控制指令的时间点、家用电器执行的控制指令、执行控制指令的家用电器的位置信息中的至少一项。
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810332006.7 | 2018-04-13 | ||
| CN201810332006.7A CN110376925A (zh) | 2018-04-13 | 2018-04-13 | 一种控制家用电器执行控制指令的方法及装置 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2019196488A1 true WO2019196488A1 (zh) | 2019-10-17 |
Family
ID=68163071
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/CN2018/121531 WO2019196488A1 (zh) | 2018-04-13 | 2019-01-09 | 一种控制家用电器执行控制指令的方法及装置 |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN110376925A (zh) |
| WO (1) | WO2019196488A1 (zh) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113541131A (zh) * | 2021-07-20 | 2021-10-22 | 西安热工研究院有限公司 | 一种火电机组子系统自动启动方法 |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112990393B (zh) * | 2019-12-02 | 2024-04-02 | 佛山市顺德区美的电热电器制造有限公司 | 操作界面的生成方法、装置 |
| CN112327648B (zh) * | 2020-11-09 | 2024-04-09 | 广东睿住智能科技有限公司 | 家用设备的控制方法、装置和计算机可读存储介质 |
| CN112286116B (zh) * | 2020-11-09 | 2023-11-07 | 扬州工业职业技术学院 | 一种基于大数据的智能家居管理系统 |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150300892A1 (en) * | 2014-04-18 | 2015-10-22 | Nest Labs, Inc. | Thermodynamic model generation and implementation using observed hvac and/or enclosure characteristics |
| CN105068515A (zh) * | 2015-07-16 | 2015-11-18 | 华南理工大学 | 一种基于自学习算法的智能家居设备语音控制方法 |
| CN105318499A (zh) * | 2015-09-30 | 2016-02-10 | 广东美的制冷设备有限公司 | 用户行为自学习空调系统及其控制方法 |
| CN106549833A (zh) * | 2015-09-21 | 2017-03-29 | 阿里巴巴集团控股有限公司 | 一种智能家居设备的控制方法和装置 |
| CN107330450A (zh) * | 2017-06-15 | 2017-11-07 | 珠海格力电器股份有限公司 | 一种终端设备控制方法和装置 |
| CN107490977A (zh) * | 2017-09-25 | 2017-12-19 | 深圳市斑点猫信息技术有限公司 | 智能家居的控制模型训练方法、控制方法及装置 |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP4123733B2 (ja) * | 2001-04-04 | 2008-07-23 | 株式会社日立製作所 | アクセス制御機器 |
| US20130297785A1 (en) * | 2012-05-04 | 2013-11-07 | Electronics And Telecommunications Research Institute | User status analyzing method and apparatus using activity history |
| CN104281113A (zh) * | 2013-07-12 | 2015-01-14 | 李文嵩 | 智能住宅系统及其运作方法 |
| CN104181898B (zh) * | 2014-09-01 | 2017-02-01 | 东北电力大学 | 一种基于分时电价响应的交互式家电智能控制方法和系统 |
| CN104865927A (zh) * | 2015-03-27 | 2015-08-26 | 北京海尔广科数字技术有限公司 | 构建用户模型的方法和装置、以及设备控制方法和装置 |
| CN105652677B (zh) * | 2016-02-24 | 2019-11-08 | 深圳台丰科技有限公司 | 一种基于用户行为分析的智能家居控制方法、装置和系统 |
-
2018
- 2018-04-13 CN CN201810332006.7A patent/CN110376925A/zh active Pending
-
2019
- 2019-01-09 WO PCT/CN2018/121531 patent/WO2019196488A1/zh active Application Filing
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150300892A1 (en) * | 2014-04-18 | 2015-10-22 | Nest Labs, Inc. | Thermodynamic model generation and implementation using observed hvac and/or enclosure characteristics |
| CN105068515A (zh) * | 2015-07-16 | 2015-11-18 | 华南理工大学 | 一种基于自学习算法的智能家居设备语音控制方法 |
| CN106549833A (zh) * | 2015-09-21 | 2017-03-29 | 阿里巴巴集团控股有限公司 | 一种智能家居设备的控制方法和装置 |
| CN105318499A (zh) * | 2015-09-30 | 2016-02-10 | 广东美的制冷设备有限公司 | 用户行为自学习空调系统及其控制方法 |
| CN107330450A (zh) * | 2017-06-15 | 2017-11-07 | 珠海格力电器股份有限公司 | 一种终端设备控制方法和装置 |
| CN107490977A (zh) * | 2017-09-25 | 2017-12-19 | 深圳市斑点猫信息技术有限公司 | 智能家居的控制模型训练方法、控制方法及装置 |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113541131A (zh) * | 2021-07-20 | 2021-10-22 | 西安热工研究院有限公司 | 一种火电机组子系统自动启动方法 |
| CN113541131B (zh) * | 2021-07-20 | 2023-09-26 | 西安热工研究院有限公司 | 一种火电机组子系统自动启动方法 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110376925A (zh) | 2019-10-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2019196488A1 (zh) | 一种控制家用电器执行控制指令的方法及装置 | |
| US10852930B2 (en) | Control method, storage medium, and information providing method | |
| EP3411634B1 (en) | Data learning server and method for generating and using learning model thereof | |
| CN105068515B (zh) | 一种基于自学习算法的智能家居设备语音控制方法 | |
| EP3539076B1 (en) | Smart assist for repeated actions | |
| WO2022041988A1 (zh) | 空调器的智能控制方法与智能控制设备 | |
| US11132547B2 (en) | Emotion recognition-based artwork recommendation method and device, medium, and electronic apparatus | |
| US20180283723A1 (en) | Data learning server and method for generating and using learning model thereof | |
| US20150286226A1 (en) | Systems and methods for updating climate control algorithms | |
| CN112789561A (zh) | 用于为电器定制便携式自然语言处理接口的系统和方法 | |
| KR20160013107A (ko) | 학습 디바이스들을 교육시키기 위한 시그널링 디바이스 | |
| CN111077786A (zh) | 基于大数据分析的智能家居设备控制方法和装置 | |
| CN110059819A (zh) | 器件的工作控制方法、装置、系统、控制设备及存储介质 | |
| CN113251610A (zh) | 用于空调控制的方法、装置和空调 | |
| CN114115027A (zh) | 目标环境参数的调节方法、系统、装置、设备及存储介质 | |
| KR102073839B1 (ko) | 스마트기기 연동 제어 시스템 및 제어 방법 | |
| US11934552B2 (en) | Selective knowledge corpus sharing | |
| CN114442697A (zh) | 一种温度控制方法、设备、介质及产品 | |
| CN113435279A (zh) | 烹饪时长的提醒方法及装置、存储介质及电子装置 | |
| CN111007731A (zh) | 用户操作预测方法、装置、电子设备及可读存储介质 | |
| KR20210006590A (ko) | 수면 품질 향상을 위한 가전기기의 맞춤형 제어 방법 및 그 제어 장치 | |
| CN112380761A (zh) | 一种基于强化学习的建筑环境控制器和控制方法 | |
| US20240122798A1 (en) | Automated temperature control system for spas | |
| US20210223748A1 (en) | Method for providing an evaluation and control model for controlling target building automation devices of a target building automation system | |
| JP6139591B2 (ja) | プログラム作成支援装置及びプログラム作成支援方法 |
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: 18914485 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: 18914485 Country of ref document: EP Kind code of ref document: A1 |