CN110824944A - Sleep behavior information prediction method and system based on intelligent household equipment - Google Patents
Sleep behavior information prediction method and system based on intelligent household equipment Download PDFInfo
- Publication number
- CN110824944A CN110824944A CN201911164246.1A CN201911164246A CN110824944A CN 110824944 A CN110824944 A CN 110824944A CN 201911164246 A CN201911164246 A CN 201911164246A CN 110824944 A CN110824944 A CN 110824944A
- Authority
- CN
- China
- Prior art keywords
- neural network
- network model
- parameter
- output
- data
- Prior art date
- Legal status (The legal status 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 status listed.)
- Pending
Links
- 230000007958 sleep Effects 0.000 title claims abstract description 97
- 230000006399 behavior Effects 0.000 title claims abstract description 70
- 238000000034 method Methods 0.000 title claims abstract description 38
- 238000003062 neural network model Methods 0.000 claims abstract description 106
- 238000012549 training Methods 0.000 claims abstract description 26
- 238000012544 monitoring process Methods 0.000 claims abstract description 6
- 238000004422 calculation algorithm Methods 0.000 claims description 20
- 238000004364 calculation method Methods 0.000 claims description 15
- 230000007613 environmental effect Effects 0.000 claims description 15
- 230000004913 activation Effects 0.000 claims description 4
- 238000004891 communication Methods 0.000 claims description 3
- 238000009795 derivation Methods 0.000 claims description 3
- 230000002123 temporal effect Effects 0.000 claims description 3
- 230000003860 sleep quality Effects 0.000 abstract description 3
- 230000006870 function Effects 0.000 description 18
- 238000013528 artificial neural network Methods 0.000 description 16
- 230000008569 process Effects 0.000 description 9
- 238000005516 engineering process Methods 0.000 description 8
- 230000000875 corresponding effect Effects 0.000 description 5
- 238000012360 testing method Methods 0.000 description 5
- 238000004458 analytical method Methods 0.000 description 4
- 238000013473 artificial intelligence Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000009434 installation Methods 0.000 description 2
- 238000012806 monitoring device Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000009776 industrial production Methods 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 230000036578 sleeping time Effects 0.000 description 1
- 230000001960 triggered effect Effects 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
- G05B19/418—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]
-
- 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
- G05B2219/00—Program-control systems
- G05B2219/20—Pc systems
- G05B2219/26—Pc applications
- G05B2219/2642—Domotique, domestic, home control, automation, smart house
Landscapes
- Engineering & Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Manufacturing & Machinery (AREA)
- Quality & Reliability (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The disclosure provides a sleep behavior information prediction method and system based on intelligent household equipment. The method comprises the following steps: monitoring the running state of one or more pieces of intelligent household equipment, and acquiring a first parameter related to the current sleep behavior when a running signal generated by triggering the intelligent household equipment is monitored; inputting the first parameter into a preset neural network model, and predicting based on the first parameter by using the neural network model so as to obtain a second parameter output after prediction; the neural network model is a model obtained by training according to historical data related to sleep behaviors; and generating the prediction information of the sleep behavior according to the second parameter. Based on the technical scheme of the invention, the sleep behavior information of the user is predicted, the sleep quality of the user is improved, and the sleep experience of the user is enhanced.
Description
Technical Field
The disclosure relates to the technical field of smart home, in particular to a sleep behavior information prediction method and system based on smart home equipment.
Background
With the development of computer technology, internet of things technology and industrial production technology, the technology is combined with conventional household products in life to produce intelligent household equipment, and the intelligent household equipment is gradually the development trend of future household equipment and products. Taking a sleep tool in an intelligent home as an example, for example, products such as a pillow and a mattress at present still mainly meet basic sleep requirements, however, a human body can generate some information in a sleep process, the information has a high value in analyzing the sleep behavior of a user, and the analysis of the information is helpful for analyzing the sleep behavior of the user, so that how to combine an artificial intelligence technology with the sleep tool to realize the analysis and prediction of the sleep behavior of the user is a research with great significance.
In the prior art, although some products have already implemented monitoring of a user sleep process, such as smart bracelets, sleep monitoring devices, and the like, the products can judge the sleep state of the user by monitoring human body vital signs, but the user can only learn the sleep condition of the user at last night or other historical periods through the products, so that analysis of the user sleep behavior information cannot be implemented, and the user sleep behavior information cannot be predicted; in addition, above-mentioned equipment such as intelligent bracelet, sleep monitoring device need the user wearing the side can use, consequently has reduced user's sleep and has experienced.
Disclosure of Invention
The invention provides a sleep behavior information prediction method and system based on intelligent home equipment, and aims to solve the problems that analysis and prediction of user sleep behavior information cannot be realized and user sleep experience is poor in related technologies.
In order to solve the technical problem, in a first aspect of the embodiments of the present disclosure, a sleep behavior information prediction method based on smart home devices is provided, including:
monitoring the running state of one or more pieces of intelligent household equipment, and acquiring a first parameter related to the current sleep behavior when a running signal generated by triggering the intelligent household equipment is monitored;
inputting the first parameter into a preset neural network model, and predicting based on the first parameter by using the neural network model so as to obtain a second parameter output after prediction; the neural network model is a model obtained by training according to historical data related to sleep behaviors;
and generating the prediction information of the sleep behavior according to the second parameter.
In some embodiments of this embodiment, the first sensor is used to monitor the operating status of one or more smart home devices, wherein,
the first sensor comprises a pressure sensor.
In some implementations of this embodiment, the first parameter includes temporal data and/or environmental data, wherein:
the time data comprises date data and time point data;
the environmental data comprises at least one of a noise value, a temperature value, a humidity value, and an air quality value;
and the noise value, the temperature value, the humidity value and the air quality value are respectively obtained through a noise sensor, a temperature sensor, a humidity sensor and an air quality sensor.
In some embodiments of this embodiment, the neural network model is a BP neural network model, and the BP neural network model is obtained by training using the following method, specifically:
acquiring historical data generated in a preset time period, wherein the historical data comprises historical time data, historical environment data and historical user sleep behavior data;
selecting a plurality of time periods from the historical data, and using a data set consisting of the time periods and the historical data corresponding to the time periods as sample data;
performing initialization operation on the structure of the BP neural network model according to the historical data to obtain an initialized BP neural network model;
and adjusting the weight and the bias in the initialized BP neural network model by utilizing an error back propagation algorithm according to the sample data so as to obtain an adjusted BP neural network model.
In some embodiments of this embodiment, adjusting the weight and the bias in the initialized BP neural network model according to the sample data and by using an error back propagation algorithm, so as to obtain an adjusted BP neural network model, includes:
inputting the sample data into an initialized BP neural network model so as to obtain actual output of the initialized BP neural network model by utilizing an activation function, a weight and bias calculation in the initialized BP neural network model;
obtaining expected output of the initialized BP neural network model, and judging the output precision of the initialized BP neural network model according to the expected output and the actual output;
when the output precision meets a preset precision condition, taking the initialized BP neural network model as an adjusted BP neural network model;
and when the output precision does not accord with the preset precision condition, adjusting the weight and the bias in the initialized BP neural network model by using an error back propagation algorithm, and enabling the output precision of the adjusted BP neural network model to be up to the preset precision condition.
In some embodiments of this embodiment, the determining the output accuracy according to the expected output and the actual output includes:
and obtaining an output error by subtracting the expected output from the actual output, comparing the output error with a preset error, judging that the output precision meets a preset precision condition when the output error is smaller than the preset error, and otherwise, judging that the output precision does not meet the preset precision condition.
In some embodiments of this embodiment, the adjusting the weights and the bias in the initialized BP neural network model by using an error back propagation algorithm includes adjusting the weights and the bias according to the following calculation formula:
wherein C (w, b) represents an error energy function; n represents the total number of training samples; x represents a training sample;
further, updating the weight in the initialized BP neural network model according to the following calculation formula, specifically:
wherein, WkRepresenting an initialization weight;
representing the partial derivative of the error energy function to the weight;
further, the bias in initializing the BP neural network model is updated according to the following calculation formula, specifically:
wherein, bιIt is shown that the initial offset is,representing the partial derivative of the error energy function to the bias;
further, the partial derivative of the error energy function to the weight and the partial derivative of the error energy function to the bias are obtained by a chain derivation rule respectively.
In some implementations of this embodiment, the generating the prediction information of the sleep behavior according to the second parameter includes:
and generating time information and environment information for predicting the sleep state of the user according to the time parameter and the environment parameter.
In some implementations of this embodiment, the smart home device includes a smart pillow and/or a smart mattress.
In a second aspect of the embodiments of the present disclosure, a sleep behavior information prediction system based on smart home devices is provided, including one or more smart home devices and a terminal device in communication connection with the smart home devices, where:
the intelligent home equipment is used for acquiring a first parameter related to the current sleep behavior when an operation signal generated by triggering the intelligent home equipment is monitored, and sending the first parameter to the terminal equipment;
the terminal device is used for receiving the first parameter, inputting the first parameter into a preset neural network model, and predicting by using the neural network model based on the first parameter so as to obtain a second parameter output after prediction; generating prediction information of the sleep behavior according to the second parameter; the neural network model is a model obtained by training according to historical data related to sleep behaviors.
The embodiment of the present disclosure adopts at least one technical scheme that can achieve the following beneficial effects:
by monitoring the operation state of one or more intelligent household devices, when an operation signal generated by triggering the intelligent household devices is monitored, a first parameter related to the current sleep behavior is obtained; inputting the first parameter into a preset neural network model, and predicting based on the first parameter by using the neural network model so as to obtain a second parameter output after prediction; the neural network model is a model obtained by training according to historical data related to sleep behaviors; and generating the prediction information of the sleep behavior according to the second parameter. Based on the technical scheme of the invention, when the user is detected to use the intelligent home equipment, the currently acquired first parameter is input into the neural network model for prediction through the neural network model trained according to the historical data related to the sleep behavior of the user, so that the information suitable for predicting the sleep state of the user is generated according to the output of the neural network model, and the prediction of the sleep behavior information of the user is realized.
Drawings
In order to more clearly illustrate the embodiments or technical solutions in the prior art of the present disclosure, the drawings used in the description of the embodiments or prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments described in the present specification, and for those skilled in the art, other drawings may be obtained according to these drawings without any creative effort.
Fig. 1 is a schematic flowchart of a sleep behavior information prediction method based on smart home devices according to an embodiment of the present disclosure;
fig. 2 is a schematic flow chart of a method for training a BP neural network model according to an embodiment of the present disclosure;
fig. 3 is a schematic flowchart of a method for adjusting weights and biases in an initialized BP neural network model according to an embodiment of the present disclosure;
fig. 4 is a schematic structural diagram of a sleep behavior information prediction system based on smart home devices according to an embodiment of the present disclosure.
Detailed Description
In order to make those skilled in the art better understand the technical solutions in the present specification, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below with reference to the drawings in the embodiments of the present specification, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the disclosed embodiments without making any creative effort, shall fall within the protection scope of the present application.
The intelligent home devices and products are various, taking sleep tools (such as pillows, mattresses and the like) in an intelligent home as an example, people can be ensured to sleep more comfortably by using the sleep tools, and the current sleep tools in the market still mainly meet the basic sleep requirements of users, so that the intelligent degree of the intelligent home devices and products is low; however, because the sleep tool is an essential household product in the sleep process, and the use frequency of the sleep tool is often regular by the user, and targeted and valuable information can be generated, the sleep tool is endowed with a more intelligent function by combining the artificial intelligence technology with the sleep tool, so that the sleep behavior of the user is analyzed, and prediction information is generated according to the sleep behavior of the user and recommended to the user, which has very important significance for assisting the user to form a good sleep behavior habit, improving the sleep quality of the user, and the like.
It should be noted that the following embodiments of the present disclosure are described with an intelligent pillow as an intelligent home device, and with relevant data generated when a user sleeps using the intelligent pillow as a processing object, and predicting a sleeping behavior of the user according to a processing result as a specific application scenario. Of course, the smart home device may further include a smart mattress, and the specific application scenario does not limit the embodiment of the disclosure.
Fig. 1 is a schematic flowchart of a sleep behavior information prediction method based on smart home devices according to an embodiment of the present disclosure, where the method specifically includes the following steps:
in step S110, an operation state of one or more smart home devices is monitored, and when an operation signal generated by triggering the smart home device is monitored, a first parameter related to a current sleep behavior is obtained.
In some embodiments of the present embodiment, the smart home device may be a smart pillow, and the operation state of the smart pillow is monitored by a first sensor installed in the smart pillow, wherein the first sensor includes a pressure sensor. In practical application, when the head of a user contacts the intelligent pillow, the intelligent pillow is triggered to generate an operation signal, which indicates that the user is ready to sleep, at the moment, a first parameter related to the current sleep behavior is acquired by using a related sensor in the intelligent pillow, and the first parameter is used as data input in actual prediction.
In some implementations of this embodiment, the first parameter includes temporal data and/or environmental data, wherein:
the time data comprises date data and time point data;
the environmental data includes at least one of a noise value, a temperature value, a humidity value, and an air quality value;
the noise value, the temperature value, the humidity value and the air quality value are respectively obtained through a noise sensor, a temperature sensor, a humidity sensor and an air quality sensor.
Further, obtaining the first parameter related to the current sleep behavior may include:
acquiring current time data; and/or the presence of a gas in the gas,
and acquiring environmental data through a second sensor, wherein the second sensor comprises at least one of a noise sensor, a temperature sensor, a humidity sensor and an air quality sensor.
Further, in some embodiments, the current time data may be obtained by a timer in the intelligent pillow, and the time data includes date data and time point data, where the date data represents a date in a calendar where the current time point is located, for example: xxx year x month x number or day of week; the point-in-time data represents the specific time currently in 24 hours, for example: 10 o' clock 30 minutes in the evening.
Further, in some embodiments, environmental data may be acquired with a second sensor within the smart pillow, such as: and acquiring related data such as a current noise value, a temperature value, a humidity value, an air quality value and the like, wherein the noise value can represent the noise of the current sleeping environment of the user, and the temperature value, the humidity value and the air quality value can represent the air condition of the current sleeping environment of the user.
It should be noted that, in the embodiment of the present disclosure, detailed limitations are not made on the specific structure of the intelligent pillow, the type and kind of the sensor, the installation position and the installation manner of the sensor, and for those skilled in the art, all manners that can achieve and ultimately achieve the corresponding effects can be applied to the embodiment of the present disclosure.
In step S120, inputting the first parameter into a predetermined neural network model, and performing prediction based on the first parameter by using the neural network model, so as to obtain a predicted and output second parameter; the neural network model is a model obtained by training according to historical data related to sleep behaviors.
In some embodiments of this embodiment, the specific structure of the neural network model includes, but is not limited to: BP neural network and its structure of variation form, full-connection neural network, RNN neural network, LSTM neural network, CNN neural network and other network structures. In the following, a training process of the BP neural network model is described in detail by taking the BP neural network model as a specific application scenario, referring to fig. 2, which shows a schematic flow chart of a method for obtaining the BP neural network model by training according to an embodiment of the present disclosure, and the BP neural network model is obtained by training with the following method, specifically:
step S210: acquiring historical data generated in a preset time period, wherein the historical data comprises historical time data, historical environment data and historical user sleep behavior data;
step S220: selecting a plurality of time periods from the historical data, and using a data set consisting of the time periods and the historical data corresponding to the time periods as sample data;
step S230: performing initialization operation on the structure of the BP neural network model according to the historical data to obtain an initialized BP neural network model;
step S240: and adjusting the weight and the bias in the initialized BP neural network model by utilizing an error back propagation algorithm according to the sample data so as to obtain an adjusted BP neural network model.
In the step S210, collecting relevant data collected by the intelligent pillow during the user sleeping process in the historical time periods of continuous days in a certain period, for example, collecting data collected by the intelligent pillow in a large number of different use environments as historical data, where the specific collection method includes, but is not limited to, collecting the operation parameters of the intelligent pillow in the laboratory simulation environment, and collecting the operation parameters of the pillow when the user actually uses the pillow through the internet of things technology.
In the step S220, the historical data is filtered, a plurality of time periods and historical data corresponding to the time periods are selected from the historical data to form a data set, and the data set is used as sample data, where the sample data includes time data, environmental data and user sleep behavior data corresponding to the time data and the environmental data, and the sample data may include a single parameter or a one-dimensional or multi-dimensional array formed by extracting features according to a certain rule.
In step S230, the basic structure of the BP neural network model, the number of input nodes, the number of output nodes, the number of hidden layers, the number of hidden layer nodes, the initial weight, the initial bias, and the like of the network may be preliminarily determined according to rules included in data of different environmental temperatures, environmental humidities, operation time periods, and the like included in the historical data.
In the step S240, the implementation process of adjusting the weight and the bias in the initialized BP neural network model by using the error back propagation algorithm according to the sample data so as to obtain the adjusted BP neural network model is described with reference to the accompanying drawings, which refer to fig. 3, which shows a schematic flow chart of a method for adjusting the weight and the bias in the initialized BP neural network model according to the embodiment of the present disclosure, specifically:
inputting sample data into the initialized BP neural network model so as to obtain actual output of the initialized BP neural network model by utilizing an activation function, a weight and bias calculation in the initialized BP neural network model;
obtaining expected output of the initialized BP neural network model, and judging the output precision of the initialized BP neural network model according to the expected output and the actual output;
when the output precision meets the preset precision condition, taking the initialized BP neural network model as the adjusted BP neural network model;
when the output precision does not meet the preset precision condition, the weight and the bias in the initialized BP neural network model are adjusted by using an Error back propagation Training algorithm (Error back propagation Training), and the output precision of the adjusted BP neural network model is enabled to be up to the preset precision condition.
Further, in some embodiments, taking the input sample data x as an example, a practical output calculation process for initializing the BP neural network model is described, specifically: inputting sample data x, calculating actual output a (x) of the network according to the activation function, the initialized weight and the bias, namely a (x) is 1/(1+ e)-z) Wherein Z ═ Wk*x+b1。
In some embodiments of this embodiment, the determining the output accuracy according to the desired output and the actual output may include:
and calculating the difference between the expected output and the actual output to obtain an output error, comparing the output error with a preset error, judging that the output precision meets the preset precision condition when the output error is smaller than the preset error, and otherwise, judging that the output precision does not meet the preset precision condition.
In a specific embodiment, for example, the expected output of the initialized BP neural network model is obtained as y (x), and whether the output precision is satisfied is determined by obtaining an output error by subtracting the expected output from the actual output, which may be determined by the following calculation formula:
| y (x) -a (x) | < ∈; e represents the target minimum error.
Further, if the output error is smaller than the target minimum error, taking the initialized BP neural network model as the adjusted BP neural network model, and finishing the training of the BP neural network model; if the output error is greater than or equal to the target minimum error, the weight and the bias in the initialized BP neural network model are adjusted by using an error back propagation algorithm by adopting the following calculation formula, specifically:
wherein C (w, b) represents an error energy function (e.g., a standard deviation function); n represents the total number of training samples; x represents a training sample;
further, updating the weight in the initialized BP neural network model according to the following calculation formula, specifically:
wherein Wk represents an initialization weight;
representing the partial derivative of the error energy function to the weight;
further, the bias in initializing the BP neural network model is updated according to the following calculation formula, specifically:
wherein, bιIt is shown that the initial offset is,
representing the partial derivative of the error energy function to the bias;
further, the partial derivative of the error energy function to the weight and the partial derivative of the error energy function to the bias are obtained by a chain derivation rule respectively.
In another embodiment of the present disclosure, for the trained BP neural network model, the forward test of the network may be continued by using the test sample, and when the test error does not meet the requirement, the above adjustment method is repeated to readjust the weight and the bias in the network until the test error meets the requirement. The obtaining mode of the test sample is the same as that of the training sample, and the description is omitted here.
Based on the content of step S120 in the above embodiment of the present disclosure, in the process of learning and training the BP neural network, the weight values between the network structure and the network nodes and the bias values of the nodes are adjusted by an error back propagation algorithm, so that the neural network fits the relationship between time data (such as time point of falling asleep and sleeping time), environmental parameters (such as temperature, humidity and air quality) and user sleeping behavior data (such as sleeping state); the sleep rule of a user using the pillow for a period of time is found through the self-learning and self-adaptive characteristics of the artificial neural network, and the relation between the past value and the future value is found, so that the curve of the user using behavior along with the change of the date and the time can be effectively fitted.
Because the theory and the performance of the BP neural network algorithm are relatively mature, compared with other traditional algorithms, the BP neural network algorithm saves a large amount of calculation and also ensures certain accuracy, and the algorithm model is adjusted only by adjusting parameters; in the later adjustment and maintenance of the traditional algorithm model, the change cost is high. The prediction is achieved by fitting a curve to the discrete points using a BP neural network, and generally, a past value is linked to a future value in one observation, the past observation value is used as an input of the BP network, and the future value is given as an output of the BP network. From the mathematical point of view, the BP neural network becomes a nonlinear function of input and output, in other words, the time prediction method based on the BP neural network is to fit the prediction function g (x) with the BP neural network and then predict the future value.
In step S130, prediction information of sleep behavior is generated according to the second parameter.
In some embodiments of this embodiment, the second parameter includes a time parameter and an environmental parameter, and the generating of the prediction information of the sleep behavior according to the second parameter may include the following:
and generating time information and environment information for predicting the sleep state of the user according to the time parameter and the environment parameter.
Taking a specific application scenario as an example, a process of generating information for predicting a sleep behavior of a user according to the second parameter and recommending the information to the user is described, specifically:
and inputting the first parameters for actual prediction into a trained BP neural network model, judging the sleep behavior rule of the user, and then generating a recommendation scheme of the user behavior according to a preset strategy. For example: judging that the user starts to fall asleep by using the pillow at 11 pm, setting 10 pm 30 minutes for the user to remind the user to fall asleep and reminding the user to turn on the air conditioner to the proper temperature if the indoor temperature and humidity of the user are proper in the sleep state according to historical data.
Further, in some embodiments, the generated prediction information may also be recommended to the user, for example, the time information and the environment information are recommended to the user, so that the system further determines the execution condition of the user on the prediction information, thereby prompting the user to adjust the sleep behavior habit of the user and improving the sleep quality of the user.
Based on the same idea, the embodiment of the present disclosure further provides a sleep behavior information prediction system based on smart home devices, including one or more smart home devices and terminal devices in communication connection with the smart home devices, wherein:
the intelligent home equipment is used for acquiring a first parameter related to the current sleep behavior when an operation signal generated by triggering the intelligent home equipment is monitored, and sending the first parameter to the terminal equipment;
the terminal device is used for receiving the first parameter, inputting the first parameter into a preset neural network model, and predicting by using the neural network model based on the first parameter so as to obtain a second parameter output after prediction; generating prediction information of the sleep behavior according to the second parameter; the neural network model is a model obtained by training according to historical data related to sleep behaviors.
In practical application, according to a specific application scenario, the terminal device may include a server and/or a mobile terminal, and the following description is given, by taking the terminal device including a server and a mobile terminal as an example, with reference to the accompanying drawings, of the structure and the function of the sleep behavior information prediction system based on the smart home device in the embodiment of the present disclosure, referring to fig. 4, which shows a schematic structural diagram of the sleep behavior information prediction system based on the smart home device provided in the embodiment of the present disclosure, and the system mainly includes:
one or more intelligent home devices 401, and a server 402 and a mobile terminal 403 connected with the intelligent home devices 401 through wireless data, wherein:
the smart home device 401 is configured to, when an operation signal generated by triggering the smart home device is monitored, acquire a first parameter related to a current sleep behavior, and send the first parameter to the server;
the server 402 is configured to receive the first parameter, input the first parameter into a predetermined neural network model, and perform prediction based on the first parameter by using the neural network model, so as to obtain a second parameter output after prediction; generating prediction information of the sleep behavior according to the second parameter; the neural network model is a model obtained by training according to historical data related to sleep behaviors;
the mobile terminal 403 is configured to send a request instruction to the smart home device and the server so as to request to acquire data of the first parameter acquired by the smart home device and to request to acquire information generated by the server and used for predicting a sleep state suitable for the user.
It should be noted that the mobile phone terminal 403 includes, but is not limited to: the Mobile phone terminal comprises a Mobile phone, a tablet computer, a wearable device and the like, wherein the Mobile phone and the tablet computer can be a smart phone or a tablet computer provided with a terminal operating system such as Syber OS, iOS, Android, Symbian, Windows Mobile, Maemo, WebOS, Palm OS or Blackberry OS, and a touch screen can be integrated on the Mobile phone terminal, and the touch screen can be a liquid crystal touch screen.
In practical application, the BP neural network algorithm model can be directly integrated in a controller of the intelligent pillow without being additionally connected with a server; the BP neural network algorithm model can also be integrated in a server, and the server is utilized to obtain data for training and prediction from the intelligent pillow.
Although the invention herein has been described with reference to particular embodiments, it is to be understood that these embodiments are merely illustrative of the principles and applications of the present invention. It is therefore to be understood that numerous modifications may be made to the illustrative embodiments and that other arrangements may be devised without departing from the spirit and scope of the present invention as defined by the appended claims. It should be understood that features described in different dependent claims and herein may be combined in ways different from those described in the original claims. It is also to be understood that features described in connection with individual embodiments may be used in other described embodiments.
Claims (10)
1. A sleep behavior information prediction method based on intelligent household equipment comprises the following steps:
monitoring the running state of one or more pieces of intelligent household equipment, and acquiring a first parameter related to the current sleep behavior when a running signal generated by triggering the intelligent household equipment is monitored;
inputting the first parameter into a preset neural network model, and predicting based on the first parameter by using the neural network model so as to obtain a second parameter output after prediction; the neural network model is a model obtained by training according to historical data related to sleep behaviors;
and generating the prediction information of the sleep behavior according to the second parameter.
2. The method of claim 1, wherein the operational status of the one or more smart home devices is monitored using a first sensor, wherein,
the first sensor comprises a pressure sensor.
3. The method of claim 1, wherein the first parameter comprises temporal data and/or environmental data, wherein:
the time data comprises date data and time point data;
the environmental data comprises at least one of a noise value, a temperature value, a humidity value, and an air quality value;
and the noise value, the temperature value, the humidity value and the air quality value are respectively obtained through a noise sensor, a temperature sensor, a humidity sensor and an air quality sensor.
4. The method of claim 1, wherein the neural network model is a BP neural network model, and the BP neural network model is trained using the following method, in particular:
acquiring historical data generated in a preset time period, wherein the historical data comprises historical time data, historical environment data and historical user sleep behavior data;
selecting a plurality of time periods from the historical data, and using a data set consisting of the time periods and the historical data corresponding to the time periods as sample data;
performing initialization operation on the structure of the BP neural network model according to the historical data to obtain an initialized BP neural network model;
and adjusting the weight and the bias in the initialized BP neural network model by utilizing an error back propagation algorithm according to the sample data so as to obtain an adjusted BP neural network model.
5. The method of claim 4, wherein adjusting weights and biases in the initialized BP neural network model according to the sample data and by using an error back propagation algorithm to obtain an adjusted BP neural network model comprises:
inputting the sample data into an initialized BP neural network model so as to obtain actual output of the initialized BP neural network model by utilizing an activation function, a weight and bias calculation in the initialized BP neural network model;
obtaining expected output of the initialized BP neural network model, and judging the output precision of the initialized BP neural network model according to the expected output and the actual output;
when the output precision meets a preset precision condition, taking the initialized BP neural network model as an adjusted BP neural network model;
and when the output precision does not accord with the preset precision condition, adjusting the weight and the bias in the initialized BP neural network model by using an error back propagation algorithm, and enabling the output precision of the adjusted BP neural network model to be up to the preset precision condition.
6. The method of claim 5, wherein determining an output accuracy based on the desired output and the actual output comprises:
and obtaining an output error by subtracting the expected output from the actual output, comparing the output error with a preset error, judging that the output precision meets a preset precision condition when the output error is smaller than the preset error, and otherwise, judging that the output precision does not meet the preset precision condition.
7. The method of claim 5, wherein adjusting weights and biases in the initialized BP neural network model using an error back propagation algorithm comprises adjusting the weights and biases according to the following calculation formula, in particular:
wherein C (w, b) represents an error energy function; n represents the total number of training samples; x represents a training sample;
further, updating the weight in the initialized BP neural network model according to the following calculation formula, specifically:
wherein, WkRepresenting an initialization weight;
representing the partial derivative of the error energy function to the weight;
further, the bias in initializing the BP neural network model is updated according to the following calculation formula, specifically:
wherein, bιIt is shown that the initial offset is,
a partial derivative of the good error energy function with respect to bias;
further, the partial derivative of the error energy function to the weight and the partial derivative of the error energy function to the bias are obtained by a chain derivation rule respectively.
8. The method of claim 1, wherein the second parameters include a time parameter and an environmental parameter, and wherein generating the prediction information of sleep behavior based on the second parameters comprises:
and generating time information and environment information for predicting the sleep state of the user according to the time parameter and the environment parameter.
9. The method of any one of claims 1-8, wherein the smart home device comprises a smart pillow and/or a smart mattress.
10. The utility model provides a sleep behavior information prediction system based on intelligent household equipment, includes one or more intelligent household equipment and with intelligent household equipment communication connection's terminal equipment, wherein:
the intelligent home equipment is used for acquiring a first parameter related to the current sleep behavior when an operation signal generated by triggering the intelligent home equipment is monitored, and sending the first parameter to the terminal equipment;
the terminal device is used for receiving the first parameter, inputting the first parameter into a preset neural network model, and predicting by using the neural network model based on the first parameter so as to obtain a second parameter output after prediction; generating prediction information of the sleep behavior according to the second parameter; the neural network model is a model obtained by training according to historical data related to sleep behaviors.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911164246.1A CN110824944A (en) | 2019-11-22 | 2019-11-22 | Sleep behavior information prediction method and system based on intelligent household equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911164246.1A CN110824944A (en) | 2019-11-22 | 2019-11-22 | Sleep behavior information prediction method and system based on intelligent household equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110824944A true CN110824944A (en) | 2020-02-21 |
Family
ID=69558849
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911164246.1A Pending CN110824944A (en) | 2019-11-22 | 2019-11-22 | Sleep behavior information prediction method and system based on intelligent household equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110824944A (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111489019A (en) * | 2020-03-31 | 2020-08-04 | 清华大学 | Sleep quality prediction method and device |
| CN111897263A (en) * | 2020-07-30 | 2020-11-06 | Oppo广东移动通信有限公司 | Intelligent glasses control method and device, storage medium and electronic equipment |
| CN112034727A (en) * | 2020-09-24 | 2020-12-04 | 广东省安心加科技有限公司 | Smart scene-based home equipment management method and smart home control device |
| CN112489720A (en) * | 2020-10-16 | 2021-03-12 | 西安得眠堂健康管理工作部 | Intelligent preparation method and device of compound peptide for improving sleep quality |
| CN113359502A (en) * | 2021-07-05 | 2021-09-07 | 信阳农林学院 | Intelligent home multi-sensor detection method and system based on artificial intelligence and storage medium |
| CN113974576A (en) * | 2021-12-23 | 2022-01-28 | 北京航空航天大学杭州创新研究院 | Sleep quality monitoring system and monitoring method based on magnetocardiogram |
| CN114063459A (en) * | 2020-08-10 | 2022-02-18 | 青岛海信电子产业控股股份有限公司 | Terminal and intelligent household control method |
| CN114065059A (en) * | 2022-01-17 | 2022-02-18 | 浙江想能睡眠科技股份有限公司 | Sleep posture recommendation control method and system based on big data and storage medium |
| WO2022142233A1 (en) * | 2020-12-28 | 2022-07-07 | 盛铭睿 | Moving bedding and model construction and sleep aiding method applied thereto, terminal, and medium |
| CN111489019B (en) * | 2020-03-31 | 2024-05-31 | 清华大学 | Sleep quality prediction method and device |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105242559A (en) * | 2015-11-04 | 2016-01-13 | 上海致能科技有限公司 | Sleep environment monitoring system and smart home system for improving sleep quality |
| CN106446552A (en) * | 2016-09-28 | 2017-02-22 | 湖南老码信息科技有限责任公司 | Prediction method and prediction system for sleep disorder based on incremental neural network model |
| CN108052014A (en) * | 2017-12-18 | 2018-05-18 | 美的集团股份有限公司 | Control method, system and the computer readable storage medium of smart home |
| CN109088806A (en) * | 2018-10-11 | 2018-12-25 | 珠海格力电器股份有限公司 | A kind of smart home sleep regulation method and system |
| CN109471370A (en) * | 2018-11-23 | 2019-03-15 | 珠海格力电器股份有限公司 | A kind of behavior prediction and control method based on exhaust fan operation data, system |
-
2019
- 2019-11-22 CN CN201911164246.1A patent/CN110824944A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105242559A (en) * | 2015-11-04 | 2016-01-13 | 上海致能科技有限公司 | Sleep environment monitoring system and smart home system for improving sleep quality |
| CN106446552A (en) * | 2016-09-28 | 2017-02-22 | 湖南老码信息科技有限责任公司 | Prediction method and prediction system for sleep disorder based on incremental neural network model |
| CN108052014A (en) * | 2017-12-18 | 2018-05-18 | 美的集团股份有限公司 | Control method, system and the computer readable storage medium of smart home |
| CN109088806A (en) * | 2018-10-11 | 2018-12-25 | 珠海格力电器股份有限公司 | A kind of smart home sleep regulation method and system |
| CN109471370A (en) * | 2018-11-23 | 2019-03-15 | 珠海格力电器股份有限公司 | A kind of behavior prediction and control method based on exhaust fan operation data, system |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111489019A (en) * | 2020-03-31 | 2020-08-04 | 清华大学 | Sleep quality prediction method and device |
| CN111489019B (en) * | 2020-03-31 | 2024-05-31 | 清华大学 | Sleep quality prediction method and device |
| CN111897263A (en) * | 2020-07-30 | 2020-11-06 | Oppo广东移动通信有限公司 | Intelligent glasses control method and device, storage medium and electronic equipment |
| CN114063459A (en) * | 2020-08-10 | 2022-02-18 | 青岛海信电子产业控股股份有限公司 | Terminal and intelligent household control method |
| CN114063459B (en) * | 2020-08-10 | 2024-03-15 | 海信集团控股股份有限公司 | Terminal and intelligent home control method |
| CN112034727A (en) * | 2020-09-24 | 2020-12-04 | 广东省安心加科技有限公司 | Smart scene-based home equipment management method and smart home control device |
| CN112489720A (en) * | 2020-10-16 | 2021-03-12 | 西安得眠堂健康管理工作部 | Intelligent preparation method and device of compound peptide for improving sleep quality |
| WO2022142233A1 (en) * | 2020-12-28 | 2022-07-07 | 盛铭睿 | Moving bedding and model construction and sleep aiding method applied thereto, terminal, and medium |
| CN113359502A (en) * | 2021-07-05 | 2021-09-07 | 信阳农林学院 | Intelligent home multi-sensor detection method and system based on artificial intelligence and storage medium |
| CN113974576B (en) * | 2021-12-23 | 2022-04-22 | 北京航空航天大学杭州创新研究院 | Sleep quality monitoring system and monitoring method based on magnetocardiogram |
| CN113974576A (en) * | 2021-12-23 | 2022-01-28 | 北京航空航天大学杭州创新研究院 | Sleep quality monitoring system and monitoring method based on magnetocardiogram |
| CN114065059A (en) * | 2022-01-17 | 2022-02-18 | 浙江想能睡眠科技股份有限公司 | Sleep posture recommendation control method and system based on big data and storage medium |
| CN114065059B (en) * | 2022-01-17 | 2022-04-19 | 浙江想能睡眠科技股份有限公司 | Sleep posture recommendation control method and system based on big data and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110824944A (en) | Sleep behavior information prediction method and system based on intelligent household equipment | |
| US10959667B2 (en) | Intelligent sleep system, and user side system and cloud side system thereof | |
| WO2017008321A1 (en) | Smart home energy management method based on smart wearable device behavior detection | |
| US11076758B2 (en) | Controlling devices based on physiological measurements | |
| Cook et al. | MavHome: An agent-based smart home | |
| US11004551B2 (en) | Sleep improvement system, and sleep improvement method using said system | |
| US7621871B2 (en) | Systems and methods for monitoring and evaluating individual performance | |
| US11268713B2 (en) | Smart home air conditioner automatic control system based on artificial intelligence | |
| JP7279445B2 (en) | Prediction method, prediction program and information processing device | |
| CN103823562A (en) | Method and system for automatically reminding users to go to bed and wearable device | |
| CN110595017B (en) | Air conditioner temperature setting method and device, computer equipment and storage medium | |
| JPWO2016157641A1 (en) | Information processing apparatus, information processing method, and program | |
| Jakkula et al. | Mining sensor data in smart environment for temporal activity prediction | |
| García et al. | System for monitoring the wellness state of people in domestic environments employing emoticon-based HCI | |
| WO2018179289A1 (en) | Area-specific environment management system, method, and program | |
| CN111012132A (en) | Sleep state adjusting method and device based on pillow and intelligent pillow | |
| CN106821349A (en) | For the event generation method and device of wearable custodial care facility | |
| KR20230144301A (en) | Encironmental health monitoring system and the method | |
| JP2014159916A (en) | Environment control apparatus, method, and program | |
| Fan et al. | A data-driven framework for thermal comfort assessment method based on user interaction | |
| CN115268269B (en) | Household energy consumption optimization system and method based on new energy low carbon | |
| US20170011184A1 (en) | Method of Adaptively Predicting Blood-Glucose Level by Collecting Biometric and Activity Data with A User Portable Device | |
| CN105893753A (en) | Health reminding method and device | |
| JP7426015B1 (en) | Healthcare data analysis system | |
| TWI816614B (en) | Sleep monitoring system and sleep monitoring method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200221 |