CN110094782A - A kind of control method and intelligent range hood of intelligent range hood - Google Patents
A kind of control method and intelligent range hood of intelligent range hood Download PDFInfo
- Publication number
- CN110094782A CN110094782A CN201910420227.4A CN201910420227A CN110094782A CN 110094782 A CN110094782 A CN 110094782A CN 201910420227 A CN201910420227 A CN 201910420227A CN 110094782 A CN110094782 A CN 110094782A
- Authority
- CN
- China
- Prior art keywords
- range hood
- signal
- intelligent range
- user
- alarm
- 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
Classifications
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24C—DOMESTIC STOVES OR RANGES ; DETAILS OF DOMESTIC STOVES OR RANGES, OF GENERAL APPLICATION
- F24C15/00—Details
- F24C15/20—Removing cooking fumes
- F24C15/2021—Arrangement or mounting of control or safety systems
Landscapes
- Engineering & Computer Science (AREA)
- Chemical & Material Sciences (AREA)
- Combustion & Propulsion (AREA)
- Mechanical Engineering (AREA)
- General Engineering & Computer Science (AREA)
- Ventilation (AREA)
- Air Conditioning Control Device (AREA)
Abstract
This application provides a kind of control method of intelligent range hood and intelligent range hood, which includes: the signal for acquiring user's body state of working region of the intelligent range hood;Determine that user's body state is normal condition or abnormality using convolutional neural networks model;According to the judging result of convolutional neural networks, it is determined whether issue the first alarm signal.Pass through the control method and intelligent range hood, it can be classified by user's body state information that convolutional neural networks will test, judge whether user is in time and the abnormalities such as fall down, if it is determined that for abnormality, then signal an alert guarantees the timely treatment or processing of user.
Description
Technical field
The present invention relates to field of kitchen electric appliance, and relate more specifically to the control method and intelligence of a kind of intelligent range hood
Kitchen ventilator.
Background technique
With the development of science and technology, the progress of social productive forces, the series of products based on artificial intelligence technology are at full speed
Stepping into the eyes of people, especially field of household appliances is even more to apply the hot fields of artificial intelligence technology, and artificial intelligence technology is answered
Huge convenience is brought with for people's lives.Constantly improve and innovating with artificial intelligence technology, many household electrical appliance
Also it is closely connect with Internet of Things-artificial intelligence technology.
Kitchen ventilator has become in kitchen appliance essential one, and conventional kitchen ventilator only has suction oil smoke, shines
The functions such as bright and display time.However, there is the old man that much lives by oneself in some families, occur in kitchen at them
In the case where falling down, household can not know in time, cause irremediable consequence.In the prior art, passing through intelligence
When energy oil smoke machine testing mobile human body, frame processing is carried out using image processing unit.In particular, image processing unit is to be imaged
It being handled based on the initial pictures of equipment acquisition, initial pictures are grayscale image, and initial pictures collected are serialized,
The initial pictures of the initial pictures and previous frame that pass sequentially through rear frame carry out the processing of frame difference, obtain the human body of each rear frame initial pictures
Moving object region.And it is primarily directed to the doing due to human body when application image processing technique detects kitchen fume concentration
Brought problem is disturbed, is not used to detection human body whether fall down in kitchen.
Therefore, it is necessary to a kind of novel intelligent kitchen ventilators, are able to detect physical condition of user's human body in kitchen, with
Family is fallen down etc. under abnormal conditions, is sounded an alarm in time, guarantees the safety treatment or processing of user.
Summary of the invention
For above-mentioned the problems of the prior art, present applicant proposes a kind of intelligent range hood and its control methods, pass through
The body state information that artificial intelligence technology will test is classified, and is judged whether user is in time and the abnormalities such as is fallen down,
If it is determined that being abnormality, then the first alarm signal is issued.
In a first aspect, the control method includes: to acquire the oil this application provides a kind of control method of intelligent range hood
The signal of user's body state in smoke machine body of work region;Based on the signal of user's body state, convolutional Neural is utilized
Network model determines that user's body state is normal condition or abnormality, and according to the judgement of convolutional neural networks model
As a result, it is determined whether issue the first alarm signal.By this aspect, the user that can be will test by convolutional neural networks model
Body state information is classified, and judge whether user is in time and the abnormalities such as is fallen down, if it is determined that for abnormality, then
Signal an alert guarantees the timely treatment or processing of user.
In one embodiment, according to the judging result of convolutional neural networks model, it is determined whether issue the first alarm
Signal specifically includes: to issue first alarm signal, the exception shape when determining user's body state is abnormality
State includes falling down state;Or when determining user's body state is normal condition, first alarm signal is not issued,
The normal condition includes walking states or sitting state.
In one embodiment, the convolutional neural networks model is trained by following steps: inputs acquisition
Multiple user's body state training signals;User's body state training signal is categorized into normal state signal or abnormality
Signal.By the embodiment, the database of user's body state signal can be pre-established, guarantees sentencing for convolutional neural networks
The accuracy of disconnected user's body state.
In one embodiment, control method further include: the electric signal of acquisition combustible hazardous gas;It is flammable based on this
The electric signal of pernicious gas calculates the corresponding concentration value of combustible hazardous gas based on the electric signal of combustible hazardous gas;In determination
When the corresponding concentration value of the combustible hazardous gas exceeds preset combustible hazardous gas concentration threshold, the second alarm signal is issued
And start suction ventilator.By the embodiment, the combustible hazardous gas being able to detect in kitchen, such as gas leakage, according to inspection
It surveys result starting suction ventilator and combustible hazardous gas is quickly discharged, guarantee the life safety of user.
In one embodiment, the control method further include: it is based on first alarm signal or the second alarm signal, to
Alarm command corresponding with the mobile terminal device transmission of kitchen ventilator subject radio connection.Pass through the embodiment, Neng Goutong
The household or friend for knowing user, indicate its further operating.
In one embodiment, control method further include: in the response instruction for receiving mobile terminal device transmission,
Stop sending the first alarm signal or the second alarm signal;Or works as and do not receive in preset duration threshold value from mobile terminal
When the response instruction of equipment, alarmed by preset warning device.By the embodiment, it can be realized and failing notice to use
In the case where family household or friend, the user in the state of falling down obtains most timely giving treatment to and handling, or makes it possible to
The leakage of processing combustible hazardous gas in time.
In one embodiment, control method further include: acquire the electric signal of oil smoke;Telecommunications based on the oil smoke
Number, calculate the corresponding concentration value of the oil smoke;In conjunction with the oil smoke concentration value that is determined by deep learning-nitrification enhancement model with
Suction ventilator is adjusted to target gear according to the corresponding concentration value of the oil smoke by the corresponding relationship of suction ventilator gear.Pass through the reality
Mode is applied, the oil smoke content being able to detect in kitchen, and the gear of suction ventilator is adaptively adjusted, reach the mesh of fume eliminating
's.
In one embodiment, the deep learning-nitrification enhancement model is trained by following steps: defeated
Enter the multiple groups training data pair of acquisition, the training data is formed to by detection target concentration value with corresponding suction ventilator tachometer value;
Classify to suction ventilator tachometer value according to gear, obtains the corresponding relationship.By the embodiment, can pre-establish flammable
The corresponding relation database of the content of pernicious gas or oil smoke and blower gear, guarantees the accuracy of classification results.
In one embodiment, the deep learning-nitrification enhancement model is DQN model or DDPG model.
Second aspect, the present invention also provides a kind of using any one of first aspect or its possible embodiment
The intelligent range hood of control method, the intelligent range hood include: the first acquisition device, are used to acquire the work of the intelligent range hood
Make user's body state in region;Controller is set to inside the intelligent range hood and is configured to:
Receive the signal of user's body state of first acquisition device transmission;Utilize the convolutional Neural built in the controller
Network model determines that user's body state is normal condition or abnormality, and according to the convolutional neural networks model
Judging result, it is determined whether issue the first alarm signal.
In one embodiment, the controller is according to the judging result of the convolutional neural networks model, determination
The first alarm signal of no sending, specifically includes: when determining user's body state is abnormality, which issues first
Alarm signal, the abnormality include falling down state;When determining user's body state is normal condition, the first alarm is not issued
Signal, the normal condition include walking states or sitting state.
In one embodiment, intelligent range hood further include: suction ventilator;Second acquisition device is used to acquire flammable
The electric signal of pernicious gas;Controller is disposed for:
Based on the electric signal of the combustible hazardous gas, the corresponding concentration value of the combustible hazardous gas is calculated;
When the concentration value for determining the combustible hazardous gas exceeds preset combustible hazardous gas concentration threshold, the is issued
Two alarm signals simultaneously start the suction ventilator.
By the embodiment, the combustible hazardous gas being able to detect in kitchen, such as gas leakage, according to testing result
Combustible hazardous gas is quickly discharged in starting suction ventilator, guarantees the life safety of user.
In one embodiment, which further includes LED light, is used to issue first in the controller alert
The flashing when number of notifying or the second alarm signal.By the embodiment, enable in combustible hazardous gas concentration be more than default
When concentration threshold, visually reminds user and shift or take other measures, avoid the further deterioration of situation;Or energy
It finds that the user in state of falling down by neighbouring personnel in time, is timely handled.
In one embodiment, which further includes receiving and dispatching with the wireless signal of controller communication connection
Device, the wireless signal transceiver are configured to:
Receive the first alarm signal or the second alarm signal of the controller;Based on received first alarm signal or
Two alarm signals, to alarm command corresponding with the mobile terminal device transmission that it is wirelessly connected.
In one embodiment, the controller is configured to: receiving the shifting in the wireless signal transceiver
When the response instruction that dynamic terminal device is sent, stop sending the first alarm signal or the second alarm signal;Or when preset
When not receiving the response instruction from the mobile terminal device in duration threshold value, alarmed by preset warning device.Pass through
The embodiment can be realized the user in the case where failing to notify user household or friend, in the state of falling down and obtain
To most timely treatment and processing, or make it possible to handle the leakage of combustible hazardous gas in time
In one embodiment, the intelligent range hood further include:
Third acquisition device is used to acquire the electric signal of oil smoke;
The controller is configured to:
Receive the electric signal of the oil smoke of third acquisition device transmission;Electric signal based on the oil smoke calculates the oil
The corresponding concentration value of cigarette;In conjunction with the oil smoke concentration value and suction ventilator gear determined by deep learning-nitrification enhancement model
The suction ventilator is adjusted to target gear according to the corresponding concentration value of the oil smoke by corresponding relationship.
In one embodiment, which is combustible hazardous gas sensor.
In one embodiment, which is oil smoke sensor.
In one embodiment, which is microwave radar, camera, pyroscan or laser thunder
One of up to.
In one embodiment, which is embedded scm.
The present invention has the advantage that compared with prior art
1) when user's body state is abnormal and (such as falls down), pass through convolutional neural networks model inspection to the exception feelings
Condition, and sound an alarm, remind the timely treatment to the user;
2) can oil smoke concentration in real-time monitoring kitchen, and adaptively automatically adjust the gear of suction ventilator, reduce oil
Smoke density;
3) intelligent range hood is communicated with mobile terminal device by wireless network transceiver, keep mobile terminal device remote
Journey receives the alarm signal, to carry out next step operation;
4) in the case where mobile terminal device fails in time to make a response the alarm signal, which can
Automatic to execute actuation of an alarm, the user for being guaranteed at abnormal (such as falling down) state timely can be handled and be given treatment to;
5) when the concentration of the combustible hazardous gas (such as coal gas) in kitchen is more than concentration threshold, suction ventilator can be started,
Combustible hazardous gas is excluded in time, guarantees the safety of user.
Above-mentioned technical characteristic may be combined in various suitable ways or be substituted by equivalent technical characteristic, as long as can reach
To the purpose of the present invention.
Detailed description of the invention
The invention will be described in more detail below based on embodiments and refering to the accompanying drawings.Wherein:
What Fig. 1 and Fig. 2 showed the control method of intelligent range hood according to an embodiment of the present invention is schematic flow chart;
Fig. 3 shows the structural schematic diagram of intelligent range hood according to another embodiment of the present invention;
Fig. 4 shows the schematic flow chart of the control method of the intelligent range hood of another embodiment according to the present invention;
Fig. 5 shows the schematic flow chart of the control method of the intelligent range hood of another embodiment according to the present invention.
Appended drawing reference:
200- intelligent range hood;210- kitchen ventilator main body;211- suction ventilator;212- oil storage box;213- keypad;214-
First acquisition device;215- wireless signal transceiver;The second acquisition device of 216-;217-LED lamp;218- third acquisition device.
In the accompanying drawings, identical component uses identical appended drawing reference.Attached drawing is not according to actual ratio.
Specific embodiment
The present invention will be further described with reference to the accompanying drawings.
Fig. 1 shows the method 100 of control intelligent range hood of the invention.As shown in Figure 1, this method 100 includes as follows
The step of:
S110 acquires the signal of user's body state in kitchen ventilator body of work region;
S120 determines that user's body state is normal condition or abnormality using convolutional neural networks model;
S130, according to the judging result of convolutional neural networks, it is determined whether issue the first alarm signal.
Wherein, S130 is specifically included:
S131 issues the first alarm signal when determining user's body state is the abnormalities such as to fall down;
S132, when determining body state is the normal conditions such as walking states or sitting state, controller does not issue first
Alarm signal.
Fig. 2 is the structural schematic diagram of intelligent range hood 200 provided by the invention.As shown in Fig. 2, the intelligent range hood 200
Including kitchen ventilator main body 210, suction ventilator 211, oil storage box 212 and keypad 213, the intelligent range hood 200 further include:
First acquisition device 214 is used to acquire the signal of user's body state of 210 working region of kitchen ventilator main body;
Controller (not shown) is set to 200 inside of intelligent range hood and is configured to:
Receive the signal of user's body state of the first acquisition device 214 transmission;
The classification that user's body state is determined using the convolutional neural networks model built in controller is normal condition or different
Normal state, and according to the judging result of convolutional neural networks, controller determines whether to issue the first alarm signal.
Wherein, when body state after determination is abnormality, controller issues the first alarm signal.
Specifically, which acquires user's human body of its working region by the first acquisition device 214 in real time
State, user's body state can be divided into normal condition and abnormality, wherein normal condition may include walking states or
Sitting state, abnormality may include the state of falling down, and user's body state signal is sent to built-in controller,
In be integrated with convolutional neural networks algorithm on the controller, the controller is in the use for receiving the transmission of the first acquisition device 214
After the body state signal of family, classification is determined to user's body state signal using convolutional neural networks algorithm, in determination
User's body state is when the abnormalities such as falling down, and controller signal an alert notifies the friend or household of user, so as to
It takes and operates in next step.
Herein, there are many factor for causing user to fall down, as kitchen surface is wet and slippery, user is old man or child, kitchen coal
Leakage or other factors for being able to use family and falling down occur for gas, and it will not be described in detail here by the present invention.
Preferably, which can be one of microwave radar, camera, pyroscan or laser radar.
It should be understood that being needed when installing intelligent range hood 200 or before using the intelligent range hood 200
The convolutional neural networks model integrated on controller in kitchen ventilator main body 210 is trained, the present invention passes through following steps
It carries out:
The multiple user's body state training signals acquired by the first acquisition device 214 are inputted to convolutional neural networks;
User's body state training signal is categorized into normal state signal or abnormality signal, wherein this is just
Normal status signal includes walking states signal and sitting status signal, and abnormality signal includes falling down status signal.
Wherein, multiple user's body state training signal can be diversified forms.For example, when the controller is microwave thunder
Up to when, which can be microwave signal, which carries the information of user's body posture on the books;And work as the control
When device processed is camera, which can be the image of its shooting.
The convolutional neural networks learn these user's body state training signals, and are classified as normal condition
Signal or abnormality signal.Since the algorithm of convolutional neural networks realizes that process is known to the skilled person
, therefore not to repeat here by the present invention.
Optionally, the first alarm signal that controller issues can be diversified forms, for example, audio signal, optical signalling or
Electronic signal.Wherein, which can be real by the audio frequency apparatus (communicating with controller) built in the intelligent range hood 210
It is existing, such as buzzer, song or one section of voice broadcast, so that nearby residents is heard and is identified;The optical signalling can pass through
The optical device (such as LED light 217) being arranged in the kitchen ventilator main body 210 is realized, is such as issued strong illumination or is continuously flashed, makes
Nearby residents are it can be seen that and identify.
Preferably, this method 100 further include:
S300 is set based on the alarm signal that controller issues to the mobile terminal being wirelessly connected with the intelligent range hood
Preparation send alarm command.
Specifically, as illustrated in fig. 2, which can also include built-in 215 (such as Wifi of wireless signal transceiver
Module).It, can be by the wireless signal transceiver 215 by corresponding first when first alarm signal is electronic signal
Alarm command sends the mobile terminal device for awarding controller communication connection, which is used to indicate user and is in different
Often (fall down) state.The mobile terminal device of remote personnel operation needs after receiving first alarm command to the nothing
Line signal transceiver 215 sends the first response instruction, and takes and operate in next step, thus reach notify long-range user friend or
The purpose of household.Wireless signal transceiver 215 stops sending the first alarm signal after receiving response instruction.Optionally, such as
When fruit does not receive the instruction of the response from mobile terminal device in preset duration threshold value, controller is filled by the alarm of threshold value
Alarm is set to ensure that the user in state of falling down timely can be given treatment to or be handled.
Preferably, in one embodiment, which further includes the second acquisition device 216, is arranged in oil
Electric signal in smoke machine main body 210 and for acquiring combustible hazardous gas.Correspondingly, as shown in figure 3, this method 100 can be with
Include:
S140 acquires the electric signal of combustible hazardous gas;
S150 calculates the corresponding concentration value of the combustible hazardous gas based on the electric signal of the combustible hazardous gas;
Whether S160 exceeds preset combustible hazardous gas concentration threshold in the corresponding concentration value for determining the combustible hazardous gas
When value, issues the second alarm signal and start suction ventilator.
That is, second acquisition device 216 acquires the electric signal of the combustible hazardous gas and the electric signal is sent to control
Device;Whether controller calculates the corresponding concentration value of the detection target, and surpass in the corresponding concentration value for determining the combustible hazardous gas
Out when preset combustible hazardous gas concentration threshold, issues the second alarm signal and start suction ventilator.
Herein, the second alarm signal can be with the first alarm signal form as described above having the same, can also
Think different forms, repeats no more herein.
Preferably, this method 100 further include:
S300 is set based on the alarm signal that controller issues to the mobile terminal being wirelessly connected with the intelligent range hood
Preparation send alarm command.
Specifically, controller sends the second alarm signal to wireless signal transceiver 215, and wireless signal transceiver 215 exists
After receiving the signal, the second alarm command is sent to the mobile terminal device being wirelessly connected with intelligent range hood, second police
Report instruction is used to indicate the leakage of combustible hazardous gas, prompts to carry out next step operation.The mobile terminal of remote personnel operation is set
It is standby to need to reply the second response of transmission to wireless transceiver 215 after receiving second alarm command and instruct, wireless signal transmitting-receiving
Device 215 stops sending the second alarm signal after receiving response instruction.Optionally, if in preset duration threshold value not
When receiving the response instruction from mobile terminal device, controller is alarmed by the warning device of threshold value.
Preferably, it is generally colourless gas in view of the combustible hazardous gas in kitchen, it is visually not noticeable, thus should
Intelligent range hood 200 further includes LED light 217, is arranged in kitchen ventilator main body 210 and for dense in the combustible hazardous gas
When degree is more than preset concentration threshold value, flashing alarm signal is issued, user is made to find the leakage of combustible hazardous gas in time.
It should be understood that the combustible hazardous gas may include coal gas.
The mobile terminal device of this paper may include that mobile phone, tablet computer or other individuals for having communication function set
Standby, this is not restricted by the present invention.
In one embodiment, which further includes third acquisition device 218, is arranged in oil smoke owner
Electric signal on body 210 and for acquiring oil smoke.Correspondingly, as shown in figure 4, this method 100 can also include:
S170 acquires the electric signal of oil smoke;
S180, the electric signal based on the oil smoke calculate the corresponding concentration value of oil smoke;
S190, in conjunction with the oil smoke concentration value and suction ventilator shelves pre-established by the powerful learning algorithm model of deep learning-
The corresponding relationship (database) of position, is adjusted to target gear for suction ventilator 211 according to the corresponding concentration value of the oil smoke.
It should be understood that needing when installing the intelligent range hood 200 or before using the intelligent range hood 200 to this
Deep learning-nitrification enhancement model is trained, and is obtained corresponding relationship (database), is carried out by following steps:
The multiple groups training data pair acquired by third acquisition device 218 is inputted, the training data is to by oil smoke concentration value
It is formed with corresponding suction ventilator tachometer value;
Classify to suction ventilator tachometer value according to gear, obtains the corresponding relationship.
For example, multiple corresponding turns of suction ventilator 211 of the acquisition of third acquisition device 218 in different oil smoke concentrations
Speed value inputs multiple groups in the deep learning-nitrification enhancement model and is made of oil smoke concentration and corresponding suction ventilator tachometer value
Data pair, the algorithm model are classified tachometer value (such as 1 grade, 2 grades and 3 grades), obtain oil smoke concentration and suction ventilator gear
Corresponding relationship, such as when oil smoke concentration value is larger, its corresponding biggish suction ventilator tachometer value is classified as 3 grades, correspondingly
When oil smoke concentration value is smaller, its corresponding lesser suction ventilator tachometer value is classified as 1 grade.In this way, in service stage,
When three acquisition devices 218 collect certain density oil smoke, controller can control suction ventilator 111 and be adjusted to corresponding gear,
Rapidly to siphon away the oil smoke.
Optionally, which can be DQN (Deep Q Network, depth Q network) algorithm
Or DDPG (Deep Deterministic Policy Gradient, depth certainty behavioral strategy) algorithm.Due to above-mentioned calculation
The realization process of method is well known to the skilled artisan, and is not repeated herein.
Multiple second acquisition devices 216 or third acquisition device 218 can be built-in in the intelligent range hood 200, it should
Second acquisition device 216 can be combustible hazardous gas sensor with the concentration signal for acquiring combustible hazardous gas, third
Acquisition device 218 can be oil smoke sensor with the concentration signal for acquiring oil smoke.
Optionally, which can be the embedded scm being arranged in inside intelligent range hood 200, embedded at this
Convolutional neural networks model and deep learning-nitrification enhancement model are integrated on single-chip microcontroller.
It should be understood that the first acquisition device 214 mentioned herein, the second acquisition device 216 and third acquisition device 218
The movement of acquisition signal can be there are many way of realization.For example, corresponding signal can be acquired in real time, it is ensured that shape is occurring
The first time of condition can find in time and make corresponding processing movement;Or can also periodically be acquired,
Such as every 1 hour, preferably 0.5 hour, the signal of acquisition in more preferably 0.1 hour, it is ensured that can handle in time.
The intelligent range hood and its control method provided through the invention can utilize artificial intelligence technology, Internet of Things skill
Art and depth learning technology improve electrical intelligence and multifunctionality, to realize: being abnormal and (such as fall in user's body state
It when), by convolutional neural networks model inspection to the abnormal conditions, and sounds an alarm, reminds the timely treatment to the user;
When the concentration of combustible hazardous gas (such as coal gas) in kitchen is more than concentration threshold, suction ventilator can be started, excluding in time can
Pernicious gas is fired, guarantees the safety of user;Can oil smoke concentration in real-time monitoring kitchen, and adaptively automatically adjust air draught
The gear of machine reduces oil smoke concentration;The intelligent range hood is communicated with mobile terminal device by wireless network transceiver, makes to move
Dynamic terminal unit remote receives the alarm signal, to carry out next step operation;Fail in time in mobile terminal device to the alarm
In the case that signal makes a response, which can execute actuation of an alarm automatically, be guaranteed at abnormal (such as falling down) shape
The user of state timely can be handled and be given treatment to.
In the description of the present invention, it is to be understood that, term " on ", "lower", "bottom", "top", "front", "rear", "inner",
The orientation or positional relationship of the instructions such as "outside", "left", "right" is to be based on the orientation or positional relationship shown in the drawings, merely to just
In description the present invention and simplify description, rather than the device or element of indication or suggestion meaning must have a particular orientation, with
Specific orientation construction and operation, therefore be not considered as limiting the invention.
Although describing the present invention herein with reference to specific embodiment, it should be understood that, these realities
Apply the example that example is only principles and applications.It should therefore be understood that can be carried out to exemplary embodiment
Many modifications, and can be designed that other arrangements, without departing from spirit of the invention as defined in the appended claims
And range.It should be understood that different appurtenances can be combined by being different from mode described in original claim
Benefit requires and feature described herein.It will also be appreciated that the feature in conjunction with described in separate embodiments can be used
In other described embodiments.
Claims (20)
1. a kind of control method of intelligent range hood, which is characterized in that the control method includes:
Acquire the signal of user's body state of the working region of the intelligent range hood;
Based on the signal of user's body state, judge that user's body state is normal using convolutional neural networks model
State or abnormality;
According to the judging result of the convolutional neural networks model, it is determined whether issue the first alarm signal.
2. control method according to claim 1, which is characterized in that according to the judgement knot of the convolutional neural networks model
Fruit, it is determined whether issue the first alarm signal, comprising:
When determining user's body state is abnormality, first alarm signal is issued, wherein the abnormality
Including the state of falling down;Or
When determining user's body state is normal condition, first alarm signal is not issued, wherein the normal shape
State includes walking states or sitting state.
3. control method according to claim 2, which is characterized in that the convolutional neural networks model passes through following steps
It is trained:
Input multiple user's body state training signals of acquisition;
User's body state training signal is categorized into normal state signal or abnormality signal.
4. control method according to claim 1, which is characterized in that the control method further include:
Acquire the electric signal of combustible hazardous gas;
Based on the electric signal of the combustible hazardous gas, the corresponding concentration value of the combustible hazardous gas is calculated;
When the corresponding concentration value for determining the combustible hazardous gas exceeds preset combustible hazardous gas concentration threshold, the is issued
Two alarm signals simultaneously start suction ventilator.
5. control method according to claim 4, which is characterized in that the control method further include:
Based on first alarm signal or second alarm signal, to the mobile end being wirelessly connected with the intelligent range hood
End equipment sends alarm command.
6. control method according to claim 5, which is characterized in that the control method further include:
When receiving the response instruction that the mobile terminal device is sent, stop sending first alarm signal or described the
Two alarm signals;Or
When not receiving the response instruction from the mobile terminal device in preset duration threshold value, pass through preset alarm
Device alarm.
7. control method according to claim 1, which is characterized in that the control method further include:
Acquire the electric signal of oil smoke;
Based on the electric signal of the oil smoke, the corresponding concentration value of the oil smoke is calculated;
In conjunction with the corresponding relationship of the oil smoke concentration value and suction ventilator gear that are determined by deep learning-nitrification enhancement model, root
Suction ventilator is adjusted to target gear according to the corresponding concentration value of the oil smoke.
8. control method according to claim 7, which is characterized in that by following steps to the deep learning-reinforcing
Learning model is trained, and obtains the corresponding relationship:
The multiple groups training data pair of acquisition is inputted, the training data is to by oil smoke concentration value and corresponding suction ventilator tachometer value group
At;
Classify to suction ventilator tachometer value according to gear, obtains the corresponding relationship.
9. control method according to claim 7 or 8, which is characterized in that the deep learning-nitrification enhancement model
For DQN model or DDPG model.
10. a kind of intelligent range hood using control method described in any one of any one of claims 1 to 99, which is characterized in that described
Intelligent range hood includes:
First acquisition device is used to acquire the signal of user's body state of the working region of the intelligent range hood;
Controller is set to inside the intelligent range hood, and is configured to:
Receive the signal for user's body state that first acquisition device is sent;
Determine that user's body state is normal condition or exception using the convolutional neural networks model built in the controller
State, and
According to the judging result of the convolutional neural networks model, it is determined whether issue the first alarm signal.
11. intelligent range hood according to claim 10, which is characterized in that the controller is according to the convolutional Neural net
The judging result of network model, it is determined whether issue the first alarm signal, specifically include:
When determining user's body state is abnormality, the controller issues the first alarm signal, wherein described different
Normal state includes falling down state;
When determining user's body state is normal condition, the controller does not issue first alarm signal, wherein
The normal condition includes walking states or sitting state.
12. intelligent range hood according to claim 10, which is characterized in that the intelligent range hood further include:
Suction ventilator;
Second acquisition device is used to acquire the electric signal of combustible hazardous gas;
Controller is disposed for:
Based on the electric signal of the combustible hazardous gas, the corresponding concentration value of the combustible hazardous gas is calculated;
When the concentration value for determining the combustible hazardous gas exceeds preset combustible hazardous gas concentration threshold, it is alert to issue second
The number of notifying simultaneously starts the suction ventilator.
13. intelligent range hood according to claim 12, which is characterized in that the intelligent range hood further includes LED light,
For being flashed when the controller issues first alarm signal or second alarm signal.
14. intelligent range hood according to claim 12, which is characterized in that the intelligent range hood further includes and the control
The wireless signal transceiver of device communication connection processed, the wireless signal transceiver are configured to:
Receive first alarm signal or second alarm signal that the controller is sent;
Based on received first alarm signal or second alarm signal, to the mobile terminal device being wirelessly connected with it
Send corresponding alarm command.
15. intelligent range hood according to claim 14, which is characterized in that the controller is configured to:
When the wireless signal transceiver receives the response instruction that the mobile terminal device is sent, stop sending described the
One alarm signal or second alarm signal;Or
When not receiving the response instruction from the mobile terminal device in preset duration threshold value, pass through preset alarm
Device alarm.
16. intelligent range hood according to claim 10, which is characterized in that the intelligent range hood further include:
Third acquisition device is used to acquire the electric signal of oil smoke;
The controller is configured to:
Receive the electric signal for the oil smoke that the third acquisition device is sent;
Based on the electric signal of the oil smoke, the corresponding concentration value of the oil smoke is calculated;
In conjunction with the corresponding relationship of the oil smoke concentration value and suction ventilator gear that are determined by deep learning-nitrification enhancement model, root
The suction ventilator is adjusted to target gear according to the corresponding concentration value of the oil smoke.
17. intelligent range hood according to claim 12, which is characterized in that second acquisition device is combustible hazardous gas
Body sensor.
18. intelligent range hood according to claim 16, which is characterized in that the third acquisition device is oil smoke sensing
Device.
19. intelligent range hood described in any one of 0 to 18 according to claim 1, which is characterized in that first acquisition device
For one of microwave radar, camera, pyroscan or laser radar.
20. intelligent range hood described in any one of 0 to 18 according to claim 1, which is characterized in that the controller is insertion
Formula single-chip microcontroller.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910420227.4A CN110094782B (en) | 2019-05-20 | 2019-05-20 | Control method of intelligent range hood and intelligent range hood |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910420227.4A CN110094782B (en) | 2019-05-20 | 2019-05-20 | Control method of intelligent range hood and intelligent range hood |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110094782A true CN110094782A (en) | 2019-08-06 |
CN110094782B CN110094782B (en) | 2020-09-04 |
Family
ID=67448648
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910420227.4A Active CN110094782B (en) | 2019-05-20 | 2019-05-20 | Control method of intelligent range hood and intelligent range hood |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110094782B (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111986417A (en) * | 2020-09-07 | 2020-11-24 | 山东臣乔电气科技股份有限公司 | Intelligent electrical fire monitoring system based on Internet of things |
CN112212380A (en) * | 2020-11-06 | 2021-01-12 | 江苏保丽洁环境科技股份有限公司 | Remote monitoring system for oil fume purifier |
CN115002956A (en) * | 2022-05-12 | 2022-09-02 | 旭宇光电(深圳)股份有限公司 | Method, device and system for intelligently and interactively controlling kitchen utensils through lamp |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
TWI250262B (en) * | 2004-09-01 | 2006-03-01 | Jing-Wen Liu | Person safety device for range hood |
JP2011232822A (en) * | 2010-04-23 | 2011-11-17 | Canon Inc | Information processing device, information processing method and program |
CN102498442A (en) * | 2009-05-19 | 2012-06-13 | 苹果公司 | Control of appliances, kitchen and home |
CN107965828A (en) * | 2017-10-30 | 2018-04-27 | 珠海格力电器股份有限公司 | The air draught condition control method and smoke exhaust ventilator of smoke exhaust ventilator |
CN108171181A (en) * | 2017-12-31 | 2018-06-15 | 南昌大学 | A kind of human fall detection method suitable in household |
CN109543607A (en) * | 2018-11-22 | 2019-03-29 | 广州小狗机器人技术有限公司 | Object abnormal state detection method, system, monitor system and storage medium |
CN109765539A (en) * | 2019-01-28 | 2019-05-17 | 珠海格力电器股份有限公司 | Indoor user behavior monitoring method and apparatus, electrical equipment and home furnishing monitoring system |
-
2019
- 2019-05-20 CN CN201910420227.4A patent/CN110094782B/en active Active
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
TWI250262B (en) * | 2004-09-01 | 2006-03-01 | Jing-Wen Liu | Person safety device for range hood |
CN102498442A (en) * | 2009-05-19 | 2012-06-13 | 苹果公司 | Control of appliances, kitchen and home |
JP2011232822A (en) * | 2010-04-23 | 2011-11-17 | Canon Inc | Information processing device, information processing method and program |
CN107965828A (en) * | 2017-10-30 | 2018-04-27 | 珠海格力电器股份有限公司 | The air draught condition control method and smoke exhaust ventilator of smoke exhaust ventilator |
CN108171181A (en) * | 2017-12-31 | 2018-06-15 | 南昌大学 | A kind of human fall detection method suitable in household |
CN109543607A (en) * | 2018-11-22 | 2019-03-29 | 广州小狗机器人技术有限公司 | Object abnormal state detection method, system, monitor system and storage medium |
CN109765539A (en) * | 2019-01-28 | 2019-05-17 | 珠海格力电器股份有限公司 | Indoor user behavior monitoring method and apparatus, electrical equipment and home furnishing monitoring system |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111986417A (en) * | 2020-09-07 | 2020-11-24 | 山东臣乔电气科技股份有限公司 | Intelligent electrical fire monitoring system based on Internet of things |
CN112212380A (en) * | 2020-11-06 | 2021-01-12 | 江苏保丽洁环境科技股份有限公司 | Remote monitoring system for oil fume purifier |
CN115002956A (en) * | 2022-05-12 | 2022-09-02 | 旭宇光电(深圳)股份有限公司 | Method, device and system for intelligently and interactively controlling kitchen utensils through lamp |
CN115002956B (en) * | 2022-05-12 | 2023-08-22 | 旭宇光电(深圳)股份有限公司 | Method, device and system for intelligent interactive control of kitchen tools through lamps |
Also Published As
Publication number | Publication date |
---|---|
CN110094782B (en) | 2020-09-04 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110094782A (en) | A kind of control method and intelligent range hood of intelligent range hood | |
KR102040505B1 (en) | Air cleaning apparatus based on artificial intelligence and air cleaning system including the same | |
CN106385457B (en) | Cloud service platform kitchen work environment intelligent warning method and system | |
CN102121906A (en) | Smoke detection system as well as intelligent control system and intelligent control method of range hood | |
CN104541106B (en) | Fume extractor and the method for controlling the fan electromotor of fan and air effect being purified for acquisition | |
CN107600789A (en) | Intelligent garbage bin | |
CN108398886B (en) | Household appliance control method, household appliance and Internet of things gateway | |
CN109799727A (en) | A kind of smart home system of long-range control curtain and window | |
KR20190096880A (en) | Artificial intelligence refrigerator | |
JP2007280167A (en) | Air conditioner | |
CN107461787A (en) | Cigarette stove linkage processing method, device, storage medium, cigarette machine and kitchen range | |
CN108344113A (en) | A kind of intelligent interconnection air cleaning system and air purification method | |
CN105159128A (en) | Intelligent household device control method and control system | |
CN110260371A (en) | Kitchen ventilator intelligence control system and method based on multisensor | |
CN104501363A (en) | Air conditioner and control system thereof | |
CN111325047A (en) | Cooking safety auxiliary method and system, kitchen appliance and combination thereof | |
CN108562088A (en) | A kind of shared refrigerator and method based on cloud platform | |
WO2023102527A1 (en) | System and method for gas detection at a field site using multiple sensors | |
CN110220232B (en) | Control method for wind speed gear of range hood and range hood | |
CN109637096A (en) | Key post personnel leave the post early warning system | |
CN106502115A (en) | A kind of intelligent home control system and method based on GPS technology | |
CN105662264A (en) | Window cleaning robot and control method thereof | |
CN214679468U (en) | Safe wisdom disinfection system | |
KR101652898B1 (en) | Apparatus and method for displaying information of sound generation | |
EP3908987B1 (en) | Method and system for reducing carbon monoxide in a building |
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 |