CN113359502A - Intelligent home multi-sensor detection method and system based on artificial intelligence and storage medium - Google Patents
Intelligent home multi-sensor detection method and system based on artificial intelligence and storage medium Download PDFInfo
- Publication number
- CN113359502A CN113359502A CN202110758950.0A CN202110758950A CN113359502A CN 113359502 A CN113359502 A CN 113359502A CN 202110758950 A CN202110758950 A CN 202110758950A CN 113359502 A CN113359502 A CN 113359502A
- Authority
- CN
- China
- Prior art keywords
- neural network
- layer
- network model
- neurons
- intelligent household
- 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.)
- Granted
Links
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], computer integrated manufacturing [CIM]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- 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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
Abstract
The invention belongs to the technical field of communication, and discloses an intelligent home multi-sensor detection method, system and storage medium based on artificial intelligence, wherein the method comprises the following steps: s1, acquiring data of a plurality of sensors on the intelligent household equipment in a working state; s2, predicting the working state data of the intelligent household equipment at the next moment by using a neural network model; s3, constructing a two-layer neural network model based on the neural network models of the intelligent household equipment; s4, predicting the cooperative work state of the intelligent household equipment at the next moment by utilizing the two-layer neural network model; s5, a control strategy for the intelligent household equipment is formulated according to the prediction result of the two-layer neural network model, and the control strategy is sent to the corresponding intelligent household equipment to control the intelligent household equipment.
Description
Technical Field
The invention belongs to the technical field of communication, and particularly relates to an intelligent home multi-sensor detection method and system based on artificial intelligence and a storage medium.
Background
Along with the development of computer technology, people's requirement to the house environment is also higher and higher, intelligent house technology comes up to the end, various equipment that intelligent house technology passes through the network in with the family is connected to together, provide household electrical appliances control, lighting control, curtain control, multiple functions such as burglar alarm, people's life style can be optimized, and is concrete, be provided with multiple sensor on the intelligent house equipment, current intelligent house control mode is, the user communicates through terminal APP and intelligent house equipment, thereby read intelligent house equipment relevant information, read sensor relevant data, and then formulate control strategy and send it to on each intelligent house equipment according to the data that acquire, thereby realize that remote control intelligent house equipment carries out autonomic work.
And less detect the intelligent home equipment under the autonomous working condition in prior art, thereby ensure that intelligent home equipment is in normal operating condition all the time, and then also guarantee the safety of family's environment, in addition, if the user wants to realize the collaborative work between a plurality of different intelligent home equipment, need set for the rule of collaborative work on the terminal APP with intelligent home equipment communication at first, for example when intelligent lock is opened, the light of entry department is opened automatically, realize the collaborative work between the different intelligent home equipment through this kind of mode, specific operation has certain complexity, and can only work according to the rule of presetting between each equipment, intelligent degree is not high, the effect of optimizing user's life style is limited.
Disclosure of Invention
The present invention is directed to solving the above problems, and provides a method, system and storage medium for intelligent home multi-sensor detection based on artificial intelligence, the invention predicts the working state of the intelligent household equipment at the next moment by acquiring the data of a plurality of sensors on the intelligent household equipment in the working state and taking the data as the input of a neural network model, then, a two-layer neural network model is constructed, and the prediction output result of the neural network model to the working state of the intelligent household equipment at the next moment is used as the input data of the two-layer neural network model to predict the cooperative working state among a plurality of intelligent household equipment at the next moment, so that the method not only can play a role in detecting whether the working state of the intelligent household equipment is normal during autonomous working, and the corresponding intelligent household equipment can be automatically controlled to carry out cooperative work according to the output result of the two-layer neural network model.
In order to achieve the above purpose, the present invention provides the following technical solution, which is achieved by the following steps:
the method comprises the steps of firstly, acquiring data of a plurality of sensors on the intelligent household equipment in a working state;
secondly, predicting the working state data of the intelligent household equipment at the next moment by using a neural network model;
thirdly, constructing a two-layer neural network model based on the neural network models of the intelligent household equipment;
fourthly, predicting cooperative working state data among the intelligent household equipment at the next moment by utilizing the two-layer neural network model;
and fifthly, formulating a control strategy for the intelligent household equipment according to the prediction result of the two-layer neural network model, and sending the control strategy to the corresponding intelligent household equipment to realize control on the intelligent household equipment.
As a preferred technical solution of the present invention, the process of predicting the working state data of the smart home device at the next time by using the neural network model in the second step is as follows:
the method comprises the steps of firstly, determining the number of neurons of an input layer and an output layer of a neural network model, specifically, determining the number of neurons of the input layer and the output layer of the neural network model depends on the respective dimensions of an input vector and an output vector of the model, wherein the number of sensors is the number of neurons of the input layer because the input vector of the model is constructed by collecting data of a plurality of sensors on the intelligent household equipment, and correspondingly, selecting the number of different working states of the intelligent household equipment as the number of neurons of the output layer;
step two, determining the number of the neurons of the hidden layer, establishing a neural network model with two hidden layers, and specifically, determining the number of the neurons of the hidden layer by using a geometric mean method
Wherein m is the number of neurons in an input layer, n is the number of neurons in an output layer, then, a neural network model with two hidden layers is established, and the neurons in different layers of the model are connected, and each connection method corresponds to one connection weight;
constructing a training data set of the neural network model, wherein the training data set not only comprises historical data of the sensor of the intelligent household equipment in a normal working state, but also comprises historical data of the sensor when the working state is abnormal;
step four, training the neural network model, specifically, initializing the connection weight of the model by a method for generating random numbers, then providing sample data for neurons in an input layer by the neural network model for each sample in a training data set, forwarding signals layer by layer until a result of an output layer is generated, calculating an error of the output layer, reversely propagating the error to neurons in a hidden layer, finally adjusting the connection weight and the error of the neurons according to the error of the neurons in the hidden layer, and stopping training the model when the training error reaches a threshold value;
and fifthly, inputting data of the multiple sensors of the intelligent household equipment at a certain moment in the working state into the neural network model, and outputting the data by the model to obtain the working state data of the corresponding intelligent household equipment at the next moment.
As a preferred technical solution of the present invention, in the second step, the neural network model is used to predict the working state data of the smart home device at the next moment, where the working state of the smart home device includes a normal working state, an abnormal working state, and a working end state.
As a preferred technical solution of the present invention, in the third step, a process of constructing a two-layer neural network model based on a plurality of neural network models of the smart home devices is as follows:
the method comprises the steps of firstly, determining the number of neurons of an input layer and an output layer of a two-layer neural network model, specifically, determining the number of neurons of the input layer and the output layer of the two-layer neural network model depends on the respective dimensions of an input vector and an output vector of the model, wherein the number of neurons of the output layer is the number of neurons of the input layer, and correspondingly, the number of cooperative work states among different intelligent household devices is the number of neurons of the output layer as the predicted output result of the neural network model to working state data of a plurality of intelligent household devices at the next moment in S2 is used as the input data of the two-layer neural network model;
step two, determining the number of the neurons of the hidden layer, establishing a neural network model with two hidden layers, and specifically, determining the number of the neurons of the hidden layer by using a geometric mean method
Wherein m is the number of neurons in an input layer, n is the number of neurons in an output layer, a neural network model with two hidden layers is established, the neurons in different layers of the model are connected, and each connection method corresponds to one connection weight;
constructing a training data set of the neural network model, wherein the training data set comprises historical state data of a sensor during cooperative work among a plurality of different intelligent household devices;
and fourthly, training the neural network model, specifically, initializing the connection weight of the model by a method for generating random numbers, providing sample data for neurons in an input layer by the neural network model for each sample in a training data set, forwarding signals layer by layer until a result of an output layer is generated, calculating an error of the output layer, reversely propagating the error to hidden neurons, adjusting the connection weight and the error of the neurons according to the error of the hidden neurons, and stopping training the model when the training error reaches a threshold value.
As a preferred technical solution of the present invention, before the second-layer neural network model is constructed based on the neural network models of the plurality of smart home devices in the third step, when the prediction output result of the neural network model in the second step is that the working state of the smart home device is about to be abnormal, a control instruction is sent to the corresponding smart home device to stop the working state.
As a preferred technical solution of the present invention, in the fourth step, the two-layer neural network model is used to predict a cooperative work state among the plurality of smart home devices at the next time, where the cooperative work state includes that different smart home devices cooperate with each other to function, and that different smart home devices perform mutual authentication.
As a preferred technical solution of the present invention, in the fifth step, a process of formulating a control strategy for the smart home device according to the prediction result of the two-layer neural network model is as follows:
firstly, verifying the accuracy of a prediction output result of a model based on the prediction output result of the cooperative working state of a plurality of intelligent household devices by combining sensor data of the intelligent household devices associated with the prediction output result in the household environment at the current moment;
step two, when the accuracy verification fails, the system generates a new training sample, adds the new training sample into a training data set corresponding to the two-layer neural network model, and trains the model again, so that the model updates the learned rule;
and step three, when the accuracy verification passes, controlling the intelligent household equipment to autonomously work according to the predicted output result of the two-layer neural network.
The invention also provides an artificial intelligence-based intelligent household multi-sensor detection system, which specifically comprises the following modules:
the intelligent home equipment comprises a first module, a second module and a third module, wherein the first module is used for acquiring data of a plurality of sensors on the intelligent home equipment in a working state;
the second module is used for constructing a neural network model and predicting the working state data of the intelligent household equipment at the next moment;
the third module is used for constructing a two-layer neural network model based on the neural network models of the plurality of intelligent household devices;
the fourth module is used for predicting cooperative work state data among the intelligent household equipment at the next moment by utilizing the two-layer neural network model;
and the fifth module is used for making a control strategy for the intelligent household equipment according to the prediction result of the two-layer neural network model and sending the control strategy to the corresponding intelligent household equipment to realize control of the intelligent household equipment.
The invention also provides a storage medium, in which instructions executable by a system of the smart home multi-sensor detection system based on artificial intelligence are stored, and the instructions are executed by a processor included in the smart home multi-sensor detection system based on artificial intelligence to realize the smart home multi-sensor detection method based on artificial intelligence.
Compared with the prior art, the invention has the following beneficial effects:
1. according to the intelligent home multi-sensor detection method based on artificial intelligence, provided by the invention, the working state of the intelligent home equipment at the next moment is predicted through a BP neural network model, so that the abnormal condition of the equipment can be detected in advance, the work of the equipment is finished in time, the problem is prevented, and the running safety of the equipment in the autonomous working state is ensured.
2. According to the artificial intelligence-based intelligent home multi-sensor detection method, the collaborative working states of the intelligent home devices are predicted through the two-layer neural network model, according to the prediction result of the model, not only can the different intelligent home devices be controlled to be mutually collaborated and play a role, but also the different intelligent home devices can be controlled to verify the working states of the intelligent home devices, the purpose of improving the intelligent degree of the intelligent home devices can be achieved, and the effect of optimizing the life style of people is obvious.
Drawings
Fig. 1 is a flowchart of a main method of an intelligent home multi-sensor detection method based on artificial intelligence according to the present invention.
Fig. 2 is a flowchart of a specific method of the intelligent home multi-sensor detection method based on artificial intelligence according to the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1, the present invention provides an intelligent home multi-sensor detection method based on artificial intelligence, which is specifically implemented by the following steps:
the method comprises the steps of firstly, acquiring data of a plurality of sensors on the intelligent household equipment in a working state.
Furthermore, a plurality of different sensors are often arranged on one intelligent household device, the sensor is a detection device, can sense measured information and convert the sensed information into an electric signal or other information output in a required form according to a certain rule so as to meet the requirements of subsequent information transmission, processing, storage, display, recording, control and the like, and is the primary link for realizing automatic detection and automatic control Acceleration sensor, electron compass sensor etc, and the data that the different moments of a plurality of sensors on the intelligent home equipment were gathered are generally different under operating condition, change along with the operating condition's of equipment change, and the sensor data of intelligent home equipment under normal operating condition is also different with the sensor data under unusual operating condition, consequently through acquireing the historical data of a plurality of sensors under operating condition, the operating condition of intelligent home equipment at next moment can be predicted, play the effect whether normal of operating condition that detects intelligent home equipment, guarantee the safety of home environment.
And secondly, predicting the working state data of the intelligent household equipment at the next moment by using the neural network model.
Further, the artificial neural network is composed of neurons, the information processing network composed of a plurality of neurons has a parallel distribution structure, each neuron has single output and can be connected with other neurons, the artificial neural network has multiple output connection methods, each connection method corresponds to a connection weight, specifically, the artificial neural network is a directed graph, when a specific problem is solved by the neural network, input data of different neurons represent different characteristics influencing problem solving, and different connection weight coefficients represent the importance degree of different characteristics on problem solving;
the prediction of the working state of the intelligent home equipment at the next moment through the BP neural network model can be regarded as a nonlinear mapping problem from the data of the multi-sensor in the working state to the working state of the equipment at the next moment, namely the input data of the model is the data of the multi-sensor on the equipment at a certain moment in the working state, and the output data is the data of the intelligent home equipment at the certain moment in the working stateThe method comprises the following steps that firstly, the number of neurons of an input layer and an output layer of a neural network model needs to be determined, specifically, the number of neurons of the input layer and the output layer depends on the respective dimensions of an input vector and an output vector of the model, the number of sensors is the number of neurons of the input layer because the input vector of the model is constructed by collecting data of a plurality of sensors on the intelligent household equipment, and correspondingly, the number of different working states of the intelligent household equipment is selected as the number of neurons of the output layer; step two, determining the number of the neurons of the hidden layer, establishing a neural network model with two hidden layers, and specifically, determining the number of the neurons of the hidden layer by using a geometric mean method
Wherein m is the number of neurons in an input layer, n is the number of neurons in an output layer, then, a neural network model with two hidden layers is established, and the neurons in different layers of the model are connected, and each connection method corresponds to one connection weight; step three, a training data set of the model is required to be established, the training data set comprises a large amount of historical data of the intelligent household equipment in a working state, for example, when the intelligent sweeping robot is performing sweeping work at a certain moment in the working state, sensor data of a plurality of moments before the moment are selected as training samples of the BP neural network model to train the model, and the training data set comprises not only the sensor historical data of the intelligent household equipment in a normal working state, but also the historical data of the sensor when the working state is abnormal; step four, initializing the connection weight of the model by a method for generating random numbers, then providing sample data for an input layer neuron by the BP neural network model for each sample in a training data set, forwarding signals layer by layer until an output layer result is generated, then calculating an error of the output layer, reversely propagating the error to a hidden layer neuron, finally adjusting the connection weight and the error of the neuron according to the error of the hidden layer neuron, stopping adjusting the connection weight and the error when the training error reaches a threshold value, and finally performing on-line multi-sensor processingData at a certain moment in the working state are input into the BP neural network model, and the working state of the corresponding intelligent household equipment at the next moment can be obtained;
in the invention, the working state of the intelligent household equipment at the next moment comprises a normal working state, an abnormal working state and a working ending state, when the output result of the BP neural network model indicates that the intelligent household equipment will continue to work in the normal state, the data of the multiple sensors at the next moment is continuously used as the input data of the BP neural network model for processing, when the output result of the BP neural network model indicates that the intelligent household equipment will end the working state, the BP neural network model is used for finishing processing the sensor data, when the output result of the BP neural network model indicates that the working state of the intelligent household equipment will be abnormal, the system sends a control instruction to the corresponding intelligent household equipment to enable the intelligent household equipment to end the working state, in conclusion, the working state of the intelligent household equipment at the next moment is predicted through the BP neural network model, the abnormal condition that equipment will take place can be detected in advance to the work of equipment is in time ended, the emergence of prevention problem, the operation safety of equipment under autonomic operating condition is guaranteed simultaneously.
And thirdly, constructing a two-layer neural network model based on a plurality of neural network models of the intelligent household equipment.
Furthermore, the two-layer neural network model takes the predicted output result of the neural network model in the second step on the working state data of the intelligent household equipment at the next moment as input data to predict the cooperative working state of the intelligent household equipment at the next moment, namely, the problem of nonlinear mapping from the working state of the intelligent household equipment at a certain moment to the cooperative working state of the intelligent household equipment at the next moment can be considered, the two-layer neural network model is constructed by using the BP neural network model, other neural network models can be selected in practice, in order to obtain the predicted result of the cooperative working state of the intelligent household equipment, the number of neurons of the input layer and the output layer of the two-layer neural network model needs to be determined in the first step, and particularly, the determination of the number of the input layer and the output neuron layer depends on the respective dimensions of the input vector and the output vector of the model, because the predicted output result of the neural network model to the working state data of the plurality of intelligent household devices at the next moment is used as the input data of the two-layer neural network model, the number of the plurality of intelligent household devices is the number of neurons of the input layer, and correspondingly, the number of the cooperative working states among different intelligent household devices is the number of neurons of the output layer; step two, determining the number of neurons of a hidden layer, establishing a neural network model with a second hidden layer, specifically, determining the number of neurons of the hidden layer by using a geometric mean method, then establishing the neural network model with the second hidden layer, connecting the neurons at different layers of the model, wherein each connection method corresponds to a connection weight; step three, a training data set of the two-layer neural network model needs to be constructed, wherein the training data set comprises historical state data of cooperative work among a plurality of different intelligent household devices, for example, when an intelligent air conditioner is in a normal working state, an intelligent door and window of a room is in a closed state, the data of the intelligent air conditioner and the intelligent door and window in the working state at the moment are selected as sample data to be used for training the two-layer neural network model; the whole process of training the two-layer neural network model in the fourth step is similar to the process of training the BP neural network model described in the second step, and is not repeated here, so that the model finally learns the internal relation between the working state of the plurality of intelligent home devices at a certain moment and the cooperative working state of the plurality of intelligent home devices at the next moment, and when the working state data of the plurality of intelligent home devices at a certain moment is input, the specific state of the cooperative working between the plurality of intelligent home devices at the next moment can be obtained through the processing of the two-layer neural network model.
And fourthly, predicting and processing cooperative working state data among the intelligent household equipment at the next moment by utilizing the two-layer neural network model.
Further, in the present invention, the working state data of the multiple smart home devices at a certain time is input into the two-layer neural network model, so as to obtain the cooperative working state data of the multiple smart home devices at the next time, where the cooperative working state data is used to formulate a control policy for the corresponding smart home devices, and specifically, the cooperative working state of the multiple smart home devices includes that different smart home devices cooperate with each other to function, and different smart home devices perform mutual authentication, and for understanding, the following describes the cooperative working state of the multiple smart home devices by way of example;
the first embodiment is as follows: the input data of the two-layer neural network model is the working state data of the intelligent air conditioner and the intelligent door and window at a certain moment, the intelligent air conditioner is in an open state, the intelligent door and window is also in an open state, and the model can predict that the working state of the intelligent door and window at the next moment is the work of automatically closing the door and window as the pre-trained two-layer neural network model learns the internal relation between the working state of the intelligent home equipment at the certain moment and the cooperative working state of the intelligent home equipment at the next moment; example two: the input data of the two-layer neural network model is the working state data of the intelligent washing machine and the intelligent clothes hanger at a certain moment, at the moment, the intelligent washing machine finishes the clothes washing work, and the intelligent clothes hanger is in a lifting state, so that the model can predict that the working state of the intelligent clothes hanger at the next moment is the action of automatic descending; the above two examples illustrate the situation that different smart home devices cooperate with each other to perform a specific function, and the following illustrates the situation that different smart home devices perform mutual authentication, example three: the input data of the second-layer neural network model is the working state data of the intelligent door and window alarm and the intelligent human body detection equipment at a certain time, at the moment, the intelligent door and window alarm is in an alarm state, and the prediction output result of the model is that the intelligent human body detection equipment is in an open state at the next time to detect whether a person intrudes into a room or not and verify the working state of the intelligent door and window alarm;
to sum up, the cooperative working state among a plurality of intelligent home devices is predicted through the two-layer neural network model, the cooperative working state among different intelligent home devices can be controlled according to the prediction result of the model, the function is exerted by controlling the different intelligent home devices to cooperate with each other, the working states among the different intelligent home devices can be controlled to be verified, the intelligent degree of the intelligent home devices can be improved, and the effect of optimizing the life style of people is obvious.
And fifthly, making a control strategy for the intelligent household equipment according to the prediction result of the two-layer neural network model, and sending the control strategy to the corresponding intelligent household equipment to realize control of the intelligent household equipment.
Further, referring to fig. 2, based on the predicted output result of the two-layer neural network model to the cooperative working state among the multiple smart home devices in the fourth step, the system makes a specific strategy for controlling the corresponding smart home devices, in order to avoid a situation that an erroneous control instruction is sent to the devices when the smart home devices are directly controlled according to the output result of the model, the accuracy of the predicted output result of the model is verified by combining sensor data of the smart home devices associated with the predicted output result in the home environment at the current time, when the verification is passed, the smart home devices are controlled to autonomously work according to the predicted result, when the verification is not passed, a new training sample is generated and added to a training data set corresponding to the two-layer neural network model, and a rule learned by the model is updated;
for example, the rule learned by the two-layer neural network model is that the air conditioner is turned on when the indoor temperature exceeds thirty degrees centigrade for more than half an hour, then when the input data of the model is that the intelligent temperature sensor detects that the indoor temperature exceeds thirty degrees centigrade for more than half an hour, and simultaneously the intelligent air conditioner is in an idle working state, the working state of the intelligent air conditioner at the next moment is predicted by the model to be the automatic starting refrigeration mode, if the intelligent human body detection sensor detects that no human exists in the room at the moment, if the intelligent air conditioner is directly controlled to be turned on according to the prediction result of the model, an error control instruction is sent to the intelligent air conditioner, the correct method under the condition is to detect that the indoor temperature exceeds thirty degrees centigrade for more than half an hour, and meanwhile, the intelligent human body detection sensor detects that a human exists in the room as a new training sample to be added into the training data set of, updating the learned rules of the model; if the intelligent human body detection sensor detects that someone is in the room, the system sends a control command for starting a refrigeration mode to the intelligent air conditioner according to a prediction result of the model, and the control command is sent to the intelligent household equipment by means of a communication technology, so that the intelligent household equipment can autonomously work according to a control strategy formulated by the system, wherein communication between the control system and the intelligent household equipment can be realized by means of a bus communication technology, a wireless communication technology, a power communication system and an Ethernet communication technology.
The invention also provides an artificial intelligence-based intelligent household multi-sensor detection system, which specifically comprises the following modules:
the intelligent home equipment comprises a first module, a second module and a third module, wherein the first module is used for acquiring data of a plurality of sensors on the intelligent home equipment in a working state;
the second module is used for constructing a neural network model and predicting the working state data of the intelligent household equipment at the next moment;
the third module is used for constructing a two-layer neural network model based on the neural network models of the plurality of intelligent household devices;
the fourth module is used for predicting cooperative work state data among the intelligent household equipment at the next moment by utilizing the two-layer neural network model;
and the fifth module is used for making a control strategy for the intelligent household equipment according to the prediction result of the two-layer neural network model and sending the control strategy to the corresponding intelligent household equipment to realize control of the intelligent household equipment.
The invention also provides a storage medium, in which instructions executable by a system of the smart home multi-sensor detection system based on artificial intelligence are stored, and the instructions are executed by a processor included in the smart home multi-sensor detection system based on artificial intelligence to realize the smart home multi-sensor detection method based on artificial intelligence.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus.
It will be appreciated by those skilled in the art that the foregoing method embodiments of the invention may be implemented as a computer program product. Thus, for example, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, optical storage, and the like) having computer-usable program code embodied therein.
Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that changes, modifications, substitutions and alterations can be made in these embodiments without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
Claims (9)
1. An intelligent home multi-sensor detection method based on artificial intelligence is characterized by comprising the following steps:
s1, acquiring data of a plurality of sensors on the intelligent household equipment in a working state;
s2, predicting the working state data of the intelligent household equipment at the next moment by using a neural network model;
s3, constructing a two-layer neural network model based on the neural network models of the intelligent household equipment;
s4, predicting cooperative working state data among the intelligent household equipment at the next moment by utilizing the two-layer neural network model;
and S5, making a control strategy for the intelligent household equipment according to the prediction result of the two-layer neural network model, and sending the control strategy to the corresponding intelligent household equipment to realize control of the intelligent household equipment.
2. The method according to claim 1, wherein the step of predicting the next working state data of the smart home device by using the neural network model in S2 is as follows:
s21, determining the number of neurons of an input layer and an output layer of a neural network model, specifically, determining the number of neurons of the input layer and the output layer depends on the respective dimensions of an input vector and an output vector of the model, because the input vector of the model is constructed by collecting data of a plurality of sensors on the intelligent household equipment, the number of the sensors is the number of the neurons of the input layer, and correspondingly, the number of different working states of the intelligent household equipment is selected as the number of the neurons of the output layer;
s22, determining the number of hidden layer neurons, establishing a neural network model with two hidden layers, and specifically, determining the number of hidden layer neurons by using a geometric mean method
Wherein m is the number of neurons in an input layer, n is the number of neurons in an output layer, then, a neural network model with two hidden layers is established, and the neurons in different layers of the model are connected, and each connection method corresponds to one connection weight;
s23, constructing a training data set of the neural network model, wherein the training data set not only comprises historical data of the sensor of the intelligent household equipment in a normal working state, but also comprises historical data of the sensor when the working state is abnormal;
s24, training the neural network model, specifically, initializing the connection weight of the model by a method for generating random numbers, providing sample data for neurons in an input layer by the neural network model for each sample in a training data set, forwarding signals layer by layer until a result of an output layer is generated, calculating an error of the output layer, reversely propagating the error to neurons in a hidden layer, adjusting the connection weight and the error of the neurons according to the error of the neurons in the hidden layer, and stopping training the model when the training error reaches a threshold value;
and S25, inputting data of the multiple sensors of the intelligent household equipment at a certain moment in the working state into the neural network model, and outputting the data by the model to obtain the working state data of the corresponding intelligent household equipment at the next moment.
3. The intelligent home multi-sensor detection method based on artificial intelligence of claim 1, wherein in S2, a neural network model is used to predict the working state data of the intelligent home device at the next moment, and the working state of the intelligent home device includes a normal working state, an abnormal working state, and a working end state.
4. The intelligent home multi-sensor detection method based on artificial intelligence of claim 1, wherein a process of constructing a two-layer neural network model based on the neural network models of the plurality of intelligent home devices in S3 is as follows:
s31, determining the number of neurons of an input layer and an output layer of a two-layer neural network model, specifically, determining the number of neurons of the input layer and the output layer of the two-layer neural network model depends on the respective dimensions of an input vector and an output vector of the model, and because the predicted output result of the neural network model to the next-moment working state data of a plurality of intelligent household devices in S2 is used as the input data of the two-layer neural network model, the number of the plurality of intelligent household devices is the number of neurons of the input layer, and correspondingly, the number of cooperative working states between different intelligent household devices is the number of neurons of the output layer;
s32, determining the number of hidden layer neurons, and establishing a hidden layer with two layersNeural network models with layers, in particular, using geometric averaging to determine the number of hidden layer neurons
Wherein m is the number of neurons in an input layer, n is the number of neurons in an output layer, a neural network model with two hidden layers is established, the neurons in different layers of the model are connected, and each connection method corresponds to one connection weight;
s33, constructing a training data set of the neural network model, wherein the training data set comprises historical state data of sensors during cooperative work among a plurality of different intelligent household devices;
s34, training the neural network model, specifically, initializing the connection weight of the model by a method for generating random numbers, providing sample data for neurons in an input layer by the neural network model for each sample in a training data set, forwarding signals layer by layer until a result of an output layer is generated, calculating an error of the output layer, reversely propagating the error to neurons in a hidden layer, adjusting the connection weight and the error of the neurons according to the error of the neurons in the hidden layer, and stopping training the model when the training error reaches a threshold value.
5. The intelligent home multi-sensor detection method based on artificial intelligence according to claim 1, wherein before a two-layer neural network model is built based on a plurality of neural network models of the intelligent home devices in S3, when a prediction output result of the neural network model in S2 indicates that an abnormal state of the intelligent home devices occurs, a control instruction is sent to the corresponding intelligent home devices to stop the working state of the intelligent home devices.
6. The method according to claim 1, wherein in S4, the two-layer neural network model is used to predict a cooperative work state among the plurality of smart home devices at the next time, where the cooperative work state includes a function of mutual cooperation among different smart home devices and a mutual authentication among different smart home devices.
7. The method according to claim 1, wherein a process of formulating a control strategy for the smart home devices according to the prediction result of the two-layer neural network model in S5 is as follows:
s51, verifying the accuracy of the prediction output result of the model by combining the prediction output result of the intelligent household equipment associated with the prediction output result in the household environment at the current moment based on the prediction output result of the cooperative working state among the intelligent household equipment of the two-layer neural network model;
s52, when the accuracy verification fails, the system generates a new training sample, adds the new training sample into a training data set corresponding to the two-layer neural network model, and trains the model again, so that the model updates the learned rule;
and S53, when the accuracy verification passes, controlling the intelligent household equipment to autonomously work according to the predicted output result of the two-layer neural network.
8. The utility model provides an intelligence house multisensor detecting system based on artificial intelligence which characterized in that includes following module:
the intelligent home equipment comprises a first module, a second module and a third module, wherein the first module is used for acquiring data of a plurality of sensors on the intelligent home equipment in a working state;
the second module is used for constructing a neural network model and predicting the working state data of the intelligent household equipment at the next moment;
the third module is used for constructing a two-layer neural network model based on the neural network models of the plurality of intelligent household devices;
the fourth module is used for predicting cooperative work state data among the intelligent household equipment at the next moment by utilizing the two-layer neural network model;
and the fifth module is used for making a control strategy for the intelligent household equipment according to the prediction result of the two-layer neural network model and sending the control strategy to the corresponding intelligent household equipment to realize control of the intelligent household equipment.
9. A storage medium having stored therein instructions executable by the system of claim 8, wherein the instructions when executed by a processor comprised by the system of claim 8 are operable to implement an artificial intelligence based intelligent home multi-sensor detection method of any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110758950.0A CN113359502B (en) | 2021-07-05 | 2021-07-05 | Intelligent household multisensor detection method, system and storage medium based on artificial intelligence |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110758950.0A CN113359502B (en) | 2021-07-05 | 2021-07-05 | Intelligent household multisensor detection method, system and storage medium based on artificial intelligence |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113359502A true CN113359502A (en) | 2021-09-07 |
| CN113359502B CN113359502B (en) | 2023-04-25 |
Family
ID=77538257
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110758950.0A Active CN113359502B (en) | 2021-07-05 | 2021-07-05 | Intelligent household multisensor detection method, system and storage medium based on artificial intelligence |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113359502B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113867167A (en) * | 2021-10-28 | 2021-12-31 | 中央司法警官学院 | Household environment intelligent monitoring method and system based on artificial neural network |
| CN114859751A (en) * | 2022-07-06 | 2022-08-05 | 深圳特科动力技术有限公司 | Intelligent household control system for realizing change regulation through centralized processing |
| CN116068910A (en) * | 2023-04-06 | 2023-05-05 | 江西财经大学 | Intelligent home control method and system based on big data |
| CN116876950A (en) * | 2023-09-05 | 2023-10-13 | 山东智赢门窗科技有限公司 | Intelligent door and window control system and method, computer equipment and storage medium |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105068515A (en) * | 2015-07-16 | 2015-11-18 | 华南理工大学 | Intelligent household equipment voice control method based on self-learning algorithm |
| CN107179700A (en) * | 2017-07-03 | 2017-09-19 | 杭州善居科技有限公司 | A kind of intelligent home control system and method based on Alljoyn and machine learning |
| CN107463106A (en) * | 2017-09-20 | 2017-12-12 | 深圳市晟达机械设计有限公司 | A kind of intelligent domestic system |
| CN107490977A (en) * | 2017-09-25 | 2017-12-19 | 深圳市斑点猫信息技术有限公司 | Controlling model training method, the control method and device of smart home |
| CN110007610A (en) * | 2019-02-25 | 2019-07-12 | 湖南悟空信息科技有限公司 | A kind of Intelligent household steward system |
| CN110084358A (en) * | 2019-03-08 | 2019-08-02 | 佛山市云米电器科技有限公司 | A kind of smart home multi-Sensor Information Fusion Approach neural network based |
| CN110824944A (en) * | 2019-11-22 | 2020-02-21 | 珠海格力电器股份有限公司 | Sleep behavior information prediction method and system based on intelligent household equipment |
| CN111880653A (en) * | 2020-07-21 | 2020-11-03 | 珠海格力电器股份有限公司 | Equipment linkage scene establishing method and device, electronic equipment and storage medium |
| CN112286110A (en) * | 2020-11-04 | 2021-01-29 | 上海美家美沪装饰科技有限公司 | Intelligent home self-adaptive interactive control system |
| CN112365074A (en) * | 2020-11-18 | 2021-02-12 | 贵州电网有限责任公司 | Artificial intelligence decision-making method based on power grid regulation and control data |
| KR102270827B1 (en) * | 2020-02-21 | 2021-06-29 | 한양대학교 산학협력단 | Generating Joint Cameraand LiDAR Features Using Cross-View Spatial Feature Mapping for 3D Object Detection |
-
2021
- 2021-07-05 CN CN202110758950.0A patent/CN113359502B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105068515A (en) * | 2015-07-16 | 2015-11-18 | 华南理工大学 | Intelligent household equipment voice control method based on self-learning algorithm |
| CN107179700A (en) * | 2017-07-03 | 2017-09-19 | 杭州善居科技有限公司 | A kind of intelligent home control system and method based on Alljoyn and machine learning |
| CN107463106A (en) * | 2017-09-20 | 2017-12-12 | 深圳市晟达机械设计有限公司 | A kind of intelligent domestic system |
| CN107490977A (en) * | 2017-09-25 | 2017-12-19 | 深圳市斑点猫信息技术有限公司 | Controlling model training method, the control method and device of smart home |
| CN110007610A (en) * | 2019-02-25 | 2019-07-12 | 湖南悟空信息科技有限公司 | A kind of Intelligent household steward system |
| CN110084358A (en) * | 2019-03-08 | 2019-08-02 | 佛山市云米电器科技有限公司 | A kind of smart home multi-Sensor Information Fusion Approach neural network based |
| CN110824944A (en) * | 2019-11-22 | 2020-02-21 | 珠海格力电器股份有限公司 | Sleep behavior information prediction method and system based on intelligent household equipment |
| KR102270827B1 (en) * | 2020-02-21 | 2021-06-29 | 한양대학교 산학협력단 | Generating Joint Cameraand LiDAR Features Using Cross-View Spatial Feature Mapping for 3D Object Detection |
| CN111880653A (en) * | 2020-07-21 | 2020-11-03 | 珠海格力电器股份有限公司 | Equipment linkage scene establishing method and device, electronic equipment and storage medium |
| CN112286110A (en) * | 2020-11-04 | 2021-01-29 | 上海美家美沪装饰科技有限公司 | Intelligent home self-adaptive interactive control system |
| CN112365074A (en) * | 2020-11-18 | 2021-02-12 | 贵州电网有限责任公司 | Artificial intelligence decision-making method based on power grid regulation and control data |
Non-Patent Citations (1)
| Title |
|---|
| 安世博等: "基于深度信念网络的家居设备状态预测模型", 《计算机技术与发展》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113867167A (en) * | 2021-10-28 | 2021-12-31 | 中央司法警官学院 | Household environment intelligent monitoring method and system based on artificial neural network |
| CN114859751A (en) * | 2022-07-06 | 2022-08-05 | 深圳特科动力技术有限公司 | Intelligent household control system for realizing change regulation through centralized processing |
| CN114859751B (en) * | 2022-07-06 | 2022-09-20 | 深圳特科动力技术有限公司 | Intelligent household control system for realizing change regulation through centralized processing |
| CN116068910A (en) * | 2023-04-06 | 2023-05-05 | 江西财经大学 | Intelligent home control method and system based on big data |
| CN116876950A (en) * | 2023-09-05 | 2023-10-13 | 山东智赢门窗科技有限公司 | Intelligent door and window control system and method, computer equipment and storage medium |
| CN116876950B (en) * | 2023-09-05 | 2023-12-05 | 山东智赢门窗科技有限公司 | Intelligent door and window control system and method, computer equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113359502B (en) | 2023-04-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113359502B (en) | Intelligent household multisensor detection method, system and storage medium based on artificial intelligence | |
| Mehr et al. | Resident activity recognition in smart homes by using artificial neural networks | |
| CN107564231B (en) | Building fire early warning and fire situation assessment system and method based on Internet of things | |
| JP6301341B2 (en) | Automatic presence detection and presence-related control in intelligent controllers | |
| CN104933841B (en) | A kind of fire prediction method based on self organizing neural network | |
| CN108983625B (en) | Intelligent household system and service generation method | |
| CN106842356B (en) | There is nobody detection method and detection system in a kind of interior | |
| CN113777938B (en) | Household appliance control method, device, equipment and storage medium | |
| CN103389710A (en) | Internet of Things environment control method and device based on context-aware | |
| CN113053053B (en) | Interval demonstration multi-sensor fusion disaster cellular alarm linkage system based on particle swarm optimization | |
| Sheikh et al. | Cyber attack and fault identification of hvac system in building management systems | |
| CN109739097A (en) | A kind of smart home robot and application thereof based on embedded type WEB | |
| CN112333142A (en) | Intelligent equipment detection system and detection method thereof | |
| Liu et al. | A Dual-spine Approach to Load Error Repair in a HEMS Sensor Network | |
| CN109243021B (en) | Deep reinforcement learning type intelligent door lock system and device based on user experience analysis | |
| Yao et al. | Establishing an ai model on data sensing and prediction for smart home environment control based on labview | |
| CN113705922B (en) | Improved ultra-short-term wind power prediction algorithm and model building method | |
| Teoh et al. | A neural network approach towards reinforcing smart home security | |
| KR101915670B1 (en) | System and Operating Method for Home Network | |
| US20230160539A1 (en) | Methods for smart gas terminal management, internet of things systems, and media thereof | |
| Zheng | Indoor smart design algorithm based on smart home sensor | |
| CN106094554A (en) | Smart Home monitoring system | |
| Zhao et al. | Analysis and design of a context-aware smart home system | |
| Boton-Fern et al. | Learning algorithm for human activity detection in smart environments | |
| Kotenko et al. | Formation of Indicators for Assessing Technical Reliability of Information Security Systems |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |