CN108536030A - A kind of intelligent domestic system and its working method based on ANFIS algorithms - Google Patents
A kind of intelligent domestic system and its working method based on ANFIS algorithms Download PDFInfo
- Publication number
- CN108536030A CN108536030A CN201810599563.5A CN201810599563A CN108536030A CN 108536030 A CN108536030 A CN 108536030A CN 201810599563 A CN201810599563 A CN 201810599563A CN 108536030 A CN108536030 A CN 108536030A
- Authority
- CN
- China
- Prior art keywords
- household electrical
- electrical appliances
- management module
- sent
- controlled 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 11
- 238000007781 pre-processing Methods 0.000 claims abstract description 24
- 238000004458 analytical method Methods 0.000 claims abstract description 7
- 230000007613 environmental effect Effects 0.000 claims description 9
- 230000006855 networking Effects 0.000 claims description 9
- 238000012549 training Methods 0.000 claims description 9
- 230000003542 behavioural effect Effects 0.000 claims description 7
- 230000003287 optical effect Effects 0.000 claims description 4
- 238000004378 air conditioning Methods 0.000 claims description 3
- 230000015572 biosynthetic process Effects 0.000 claims description 3
- 230000008859 change Effects 0.000 claims description 3
- 238000009434 installation Methods 0.000 claims description 3
- 230000001932 seasonal effect Effects 0.000 claims description 3
- 230000002123 temporal effect Effects 0.000 claims description 3
- 239000004984 smart glass Substances 0.000 claims 1
- 238000001514 detection method Methods 0.000 abstract 1
- 230000006399 behavior Effects 0.000 description 19
- 230000006870 function Effects 0.000 description 17
- 230000005284 excitation Effects 0.000 description 9
- 238000005286 illumination Methods 0.000 description 8
- 230000003044 adaptive effect Effects 0.000 description 5
- 238000013528 artificial neural network Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 238000013507 mapping Methods 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 238000013461 design Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000013213 extrapolation Methods 0.000 description 2
- 238000007792 addition Methods 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000012141 concentrate Substances 0.000 description 1
- 230000001186 cumulative effect Effects 0.000 description 1
- 238000012217 deletion Methods 0.000 description 1
- 230000037430 deletion Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 238000007667 floating Methods 0.000 description 1
- 238000012886 linear function Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000001537 neural effect Effects 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B15/00—Systems controlled by a computer
- G05B15/02—Systems controlled by a computer electric
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B19/00—Programme-control systems
- G05B19/02—Programme-control systems electric
- G05B19/418—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS] or computer integrated manufacturing [CIM]
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B2219/00—Program-control systems
- G05B2219/20—Pc systems
- G05B2219/26—Pc applications
- G05B2219/2642—Domotique, domestic, home control, automation, smart house
-
- 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]
Landscapes
- Engineering & Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Manufacturing & Machinery (AREA)
- Quality & Reliability (AREA)
- Selective Calling Equipment (AREA)
Abstract
The present invention relates to a kind of intelligent domestic system and its working method based on ANFIS algorithms, belongs to technical field of intelligent home control.The present invention's includes node device module, network management module, intelligent gateway, user management module；Node device module is connect with network management module, and network management module is connect with intelligent gateway, and intelligent gateway is connect with user management module；User sends control instruction by user management module first, toy data base has user's control instruction by storage, sensor device detection information, time data, the room number of controlled household electrical appliances, device number is sent to data preprocessing module, data preprocessing module goes out and the relevant data of controlled household electrical appliances according to the information analysis of reception, user behavior custom prediction model obtains the use habit of user according to the output signal of data preprocessing module, it does not need user and sends control instruction, user behavior custom prediction model can directly control the work of internal control household electrical appliances, intelligent height.
Description
Technical field
The present invention relates to a kind of intelligent domestic system and its working method based on ANFIS algorithms, belongs to smart home control
Technical field processed.
Technical background
Since stepping into 21 century, science and technology is grown rapidly human society, and people’s lives level is continuously improved.Full
While foot basic living conditions, people are to the quality of the life of family, such as the comfort level of family life, intelligence, safety
Etc. there is higher pursuit, the requirement to science and technology is improved there has also been further.Wherein, wireless sensor technology and computer technology
Development allow people that can enjoy more convenient, comfortable, intelligent life, traditional domestic environment can not meet people increasingly
The demand of growth, smart home arise.
But the concept of smart home design at present and is matched often by wired or wireless network access appliance equipment
It covers corresponding cell phone application or the ability of remote control is realized at the ends Web.Though remote monitoring and modification household electric may be implemented in user
The ability of device equipment, but this has only accomplished the remote control of " artificial trigger-type ", for a certain degree, such long-range control
System can not can be regarded as intelligent, automation.
Invention content
The technical problem to be solved in the invention is to provide a kind of intelligent domestic system based on ANFIS algorithms and its work
Method can make in the case where not needing user and actively manipulating household appliance according to the living habit analog subscriber of user
Corresponding operation has the function that manage home environment.
The technical solution adopted by the present invention is：A kind of intelligent domestic system based on ANFIS algorithms, including node device mould
Block, network management module, intelligent gateway, user management module；
Node device module is connect with network management module, and network management module is connect with intelligent gateway, intelligent gateway with
Family management module connection；
User management module sends the control instruction that control is controlled household electrical appliances for user；
Node device module includes sensor device, controlled household electrical appliances；
Network management module is used to, by sensor and controlled household electrical appliances wireless networking, obtain the room number for being controlled household electrical appliances, equipment
Number, and the environmental information that the room number, device number and sensor device of controlled household electrical appliances detect is sent to intelligent gateway, simultaneously
Control instruction for sending intelligent gateway is sent to controlled household electrical appliances；
Intelligent gateway includes that controller, toy data base, data preprocessing module and user behavior are accustomed to prediction model, small
Type database is connect with and with controller, data preprocessing module respectively, controller simultaneously with user management module, network management
Module connects；Controller is sent respectively to toy data base, network management mould for user management module to be sent control instruction
Block, and room number, the device number of the controlled household electrical appliances that network management module is sent are sent to toy data base, toy data base
Time data for storing each controlled household electrical appliances, and for receiving and storing control instruction, the sensing that user management module is sent
The room number and device number for the controlled household electrical appliances that information that device device is sent, network management module are sent, and by the information of storage
It is sent to data preprocessing module, data preprocessing module is used to receive the information that toy data base is sent and by determining data
The algorithm of correlation filters out the data larger with controlled household electrical appliances working condition correlation, is then passed to user behavior custom prediction
Model；User behavior custom prediction model is used to go out by ANFIS network analyses the behavioural habits of user, then forms control quilt
The control instruction of household electrical appliances is controlled, and the control instruction of formation is sent to controlled household electrical appliances by network management module.
Preferably, the algorithm of the determination data dependence includes single input ANFIS algorithms.
Preferably, the sensor device includes Temperature Humidity Sensor, optical sensor, controlled household electrical appliances include air-conditioning,
Humidifier, intelligent electric lamp, intelligent curtain.
Preferably, the network management module uses ZigBee wireless networkings, the ZigBee models CC2530 of use.
Preferably, the user management module is the ends PC or cell phone application end or Web page, and toy data base uses
SQLite3。
Preferably, the time data of the controlled household electrical appliances include controlled household electrical appliances use time, season, month, week,
Specific time, the control instruction that user management module is sent refer to the on off state of controlled household electrical appliances.
Preferably, the time is the time for generating current data, is preserved with int types；
Season is to distinguish the on off state of the controlled household electrical appliances of Various Seasonal in 1 year, and numberical range is that the int of 1-4 is whole
Type；
It is to discriminate between the slight change that appliance switch state is controlled in the same season month, numberical range is the int of 1-12
Type；
Week is to distinguish the specific works state for time for being controlled household electrical appliances in a calendar month, and numberical range is 1-7's
Int types；
The specific time is the specific works state for time that household electrical appliances are controlled in 24 hours one day, and time interval is to remember for every 10 minutes
Record is primary, and numerical value saves as int types；
Room number is to discriminate between the region of controlled household electrical appliances installation, and numberical range is the sum of the rooms 1- number；
Device number is to discriminate between the unique mark of different controlled household electrical appliances, and numberical range is the sum of 1- number of devices；
Equipment state is to show the working condition of current controlled household electrical appliances, and 0 expression current device is closed, and 1 indicates
Current device is in running order.
A kind of working method of the intelligent domestic system based on ANFIS algorithms, includes the following steps：
step1:Control command is sent to the controller in intelligent gateway by user by user management module, and controller will
Control instruction is sent to toy data base storage, while control instruction is sent to node device module by network management module
In controlled household electrical appliances；
step2:Node device module acquires indoor environment Data Concurrent by sensor device and gives network management module；
step3：Network management module realizes sensor device and controlled household electrical appliances in family by ZigBee wireless
Networking obtains room number, the device number for being controlled household electrical appliances, and by room number, device number and the sensor of the controlled household electrical appliances of acquisition
The environmental information that device detects is sent to the controller in intelligent gateway；
step4：The information that controller receiving network managing module in intelligent gateway is sent, and the information of reception is sent
To toy data base, toy data base itself is stored with the temporal information of controlled household electrical appliances, toy data base by controlled household electrical appliances when
Between information, room number, device number and controller control instruction, the environmental information of sensor device of user management module sent
It is sent to data preprocessing module, data preprocessing module is by the information received by determining that the algorithm of data dependence screens
Go out the data larger with controlled household electrical appliances working condition correlation, is then passed to user behavior custom prediction model；
step5：User behavior custom prediction model goes out the behavioural habits of user by ANFIS network analyses, is then formed
Control is controlled the control instruction of household electrical appliances, and control instruction is automatically transmitted to controlled household electrical appliances by network management module.
Specifically, in the step4, determine that the algorithm of data dependence includes the following steps：
step4.1：Build single input ANFIS：Input variable is combined with output variable respectively, obtains two tuple (x1,
Y), (x2, y) ..., (xn, y), two tuples are inputted into single input ANFIS, training network is learnt, and prediction result is obtained
And corresponding prediction error e can be obtained1, e2..., en,
Wherein, xiI-th of input vector x is indicated respectively；Y is system output, eiFor i-th factor prediction of input vector x
The prediction error of value and actual value, i=1,2 ..., n；
step4.2：The percentage that each prediction error accounts for overall error is calculated, formula is as follows：
Wherein,The percentage of overall error is accounted for for each prediction error；
step4.3：Given threshold, and each of will calculate percentage and threshold value ratio that prediction error accounts for overall error
Compared with, cast out prediction error account for overall error percentage be more than given threshold input variable；
step4.4：It is accustomed to the input of prediction model using the input variable preserved as user behavior.
The beneficial effects of the present invention are：
(1) system has good learning ability, adaptive ability and self organization ability：System by machine learning algorithm and
Smart home is combined, and ANFIS systems are embedded in its intelligent domestic gateway, this is by T-S types fuzzy inference system and manually
Neural network is combined, which has the advantages that fuzzy inference system and artificial neural network respectively, has powerful self-study
Habit ability and nonlinear extrapolation characteristic.
(2) system can make decisions according to the living habit analog subscriber of user：ANFIS systems have good study
Ability, the system can help the environment configurations of the good family of user management according to the living habit of user in user stays out,
It realizes really intelligent.
(3) system uses modularization programming, favorable expandability：The various pieces of the system are all made of modularization programming, are convenient for
The maintenance of user's additions and deletions node device and equipment, simultaneity factor employing wireless sensing network are not necessarily to cumbersome wiring.System uses
The small-sized databases of SQLite3, are embedded into intelligent gateway by intelligent algorithm, as the part of data processing and aid decision, subtract
Small EMS memory occupation.
(4) there is smaller training time and error：Improved ANFIS systems use is trained in batches, only training with it is defeated
Go out premise parameter and consequent parameter caused by the related input item of item, so the parameter of training is less, the time is shorter.
Description of the drawings
Fig. 1 is a kind of structure diagram of the intelligent domestic system based on ANFIS algorithms of the present invention；
Fig. 2 is a kind of flow diagram of the intelligent domestic system based on ANFIS algorithms of the present invention；
Fig. 3 is a kind of flow chart of the working method of the intelligent domestic system based on ANFIS algorithms of the present invention；
Fig. 4 is the layer of structure figure of ANFIS systems of the present invention；
Fig. 5 is a kind of algorithm flow chart of determining data dependence of the present invention.
Specific implementation mode
In the following with reference to the drawings and specific embodiments, the invention will be further elaborated.
Embodiment 1：As shown in Figs. 1-5, a kind of intelligent domestic system based on ANFIS algorithms, including node device module,
Network management module, intelligent gateway, user management module；
Node device module is connect with network management module, and network management module is connect with intelligent gateway, intelligent gateway with
Family management module connection；
User management module sends the control instruction that control is controlled household electrical appliances for user；
Node device module includes sensor device, controlled household electrical appliances；
Network management module is used to, by sensor and controlled household electrical appliances wireless networking, obtain the room number for being controlled household electrical appliances, equipment
Number, and the environmental information that the room number, device number and sensor device of controlled household electrical appliances detect is sent to intelligent gateway, simultaneously
Control instruction for sending intelligent gateway is sent to controlled household electrical appliances；
Intelligent gateway includes that controller, toy data base, data preprocessing module and user behavior are accustomed to prediction model, small
Type database is connect with and with controller, data preprocessing module respectively, controller simultaneously with user management module, network management
Module connects；Controller is sent respectively to toy data base, network management mould for user management module to be sent control instruction
Block, and room number, the device number of the controlled household electrical appliances that network management module is sent are sent to toy data base, toy data base
Time data for storing each controlled household electrical appliances, and for receiving and storing control instruction, the sensing that user management module is sent
The room number and device number for the controlled household electrical appliances that information that device device is sent, network management module are sent, and by the information of storage
It is sent to data preprocessing module, data preprocessing module is used to receive the information that toy data base is sent and by determining data
The algorithm of correlation filters out the data larger with controlled household electrical appliances working condition correlation, is then passed to user behavior custom prediction
Model；User behavior custom prediction model is used to go out by ANFIS network analyses the behavioural habits of user, then forms control quilt
The control instruction of household electrical appliances is controlled, and the control instruction of formation is sent to controlled household electrical appliances by network management module.
Further, the algorithm of the determination data dependence includes single input ANFIS algorithms.
Further, the sensor device includes Temperature Humidity Sensor, optical sensor, and Temperature Humidity Sensor is adopted
With DHT11 models；Optical sensor uses GY-30 models；Controlled household electrical appliances include air-conditioning, humidifier, intelligent electric lamp, intelligent curtain
Deng.
Further, the network management module uses ZigBee wireless networkings, the ZigBee models of use
CC2530。
Further, the user management module is the ends PC or cell phone application end or Web page, and toy data base uses
SQLite3。
Further, the time data of the controlled household electrical appliances include controlled household electrical appliances use time, season, month, star
Phase, specific time, the control instruction that user management module is sent refer to the on off state of controlled household electrical appliances.
Further, the time is the time for generating current data, is preserved with int types；
Season is to distinguish the on off state of the controlled household electrical appliances of Various Seasonal in 1 year, and numberical range is that the int of 1-4 is whole
Type；
It is to discriminate between the slight change that appliance switch state is controlled in the same season month, numberical range is the int of 1-12
Type；
Week is to distinguish the specific works state for time for being controlled household electrical appliances in a calendar month, and numberical range is 1-7's
Int types；
The specific time is the specific works state for time that household electrical appliances are controlled in 24 hours one day, and time interval is to remember for every 10 minutes
Record is primary, and numerical value saves as int types；
Room number is to discriminate between the region of controlled household electrical appliances installation, and numberical range is the sum of the rooms 1- number；
Device number is to discriminate between the unique mark of different controlled household electrical appliances, and numberical range is the sum of 1- number of devices；
Equipment state is to show the working condition of current controlled household electrical appliances, and 0 expression current device is closed, and 1 indicates
Current device is in running order.
Temperature, humidity, intensity of illumination are to describe the design parameter of home environment, and correlation is acquired by different sensors
Data, numerical value saves as float type floating numbers.
A kind of working method of the intelligent domestic system based on ANFIS algorithms, includes the following steps：
step1:Control command is sent to the controller in intelligent gateway by user by user management module, and controller will
Control instruction is sent to toy data base storage, while control instruction is sent to node device module by network management module
In controlled household electrical appliances；
step2:Node device module acquires indoor environment Data Concurrent by sensor device and gives network management module；
step3：Network management module realizes sensor device and controlled household electrical appliances in family by ZigBee wireless
Networking obtains room number, the device number for being controlled household electrical appliances, and by room number, device number and the sensor of the controlled household electrical appliances of acquisition
The environmental information that device detects is sent to the controller in intelligent gateway；
step4：The information that controller receiving network managing module in intelligent gateway is sent, and the information of reception is sent
To toy data base, toy data base itself is stored with the temporal information of controlled household electrical appliances, toy data base by controlled household electrical appliances when
Between information, room number, device number and controller control instruction, the environmental information of sensor device of user management module sent
It is sent to data preprocessing module, data preprocessing module is by the information received by determining that the algorithm of data dependence screens
Go out the data larger with controlled household electrical appliances working condition correlation, is then passed to user behavior custom prediction model；
step5：User behavior custom prediction model goes out the behavioural habits of user by ANFIS network analyses, is then formed
Control is controlled the control instruction of household electrical appliances, and control instruction is automatically transmitted to controlled household electrical appliances by network management module.
Certainly, user can send the working condition that control instruction control is controlled household electrical appliances in real time by user management module.
Specifically, in the step4, the algorithm of data dependence is determined as shown in figure 5, based on ANFIS, is come true
The functional relation of the power and input data and output result of relevance of the fixed number between,
The algorithm includes the following steps：
step4.1：Build single input ANFIS：Input variable is combined with output variable respectively, obtains two tuple (x1,
Y), (x2, y) ..., (xn, y), two tuples are inputted into single input ANFIS, training network is learnt, and prediction result is obtained
And corresponding prediction error e can be obtained1, e2..., en,
Wherein, xiI-th of input vector x is indicated respectively；Y is system output, eiFor i-th factor prediction of input vector x
The prediction error of value and actual value, i=1,2 ..., n；
step4.2：The percentage that each prediction error accounts for overall error is calculated, formula is as follows：
Wherein,The percentage of overall error is accounted for for each prediction error；
step4.3：Given threshold, and each of will calculate percentage and threshold value ratio that prediction error accounts for overall error
Compared with, cast out prediction error account for overall error percentage be more than given threshold input variable；
step4.4：It is accustomed to the input of prediction model using the input variable preserved as user behavior.
The present invention acquires home environment signal by node device module, and collected signal concentration is sent to wirelessly
Network module；Intelligent gateway will receive from wireless network module and concentrate the data sent, and to these data processings and be shown in
The ends Web or mobile phone terminal；Intelligent gateway records input of the data of user setting as data preprocessing module, obtains and control household electrical appliances
The larger data of working condition correlation, the ANFIS systems being then input in user behavior custom prediction model, pass through
The processing of ANFIS systems, system can predict that the custom of user, user need not manipulate household appliance in person, and system being capable of root
Carry out analog subscriber according to the custom of user and manipulates household appliance, meanwhile, user can also be according to the wish and demand control quilt of oneself
Control object.
Adaptive neuro-fuzzy inference system (ANFIS) is by T-S types fuzzy inference system and artificial neural network phase
In conjunction with the system has the advantages that fuzzy inference system and artificial neural network respectively, and maximum feature is according to datum
According to can be modeled to control system；The system has powerful self-learning capability and nonlinear extrapolation characteristic, essence
It is to realize linearly or nonlinearly mapping of the input variable to output variable by learning training data, and obtaining mapping pass
The estimated value of output variable is provided after system.
As shown in figure 3, when user's control household appliance, intelligent gateway first carry out order and record order be stored in it is small-sized
In database, after data preprocessing module extracts the data of toy data base storage, with a kind of algorithm of determining data dependence
Different data cluster is pre-processed, different types of data, such as time, season, week different factors is distinguished, distinguishes different
The data in region, while the correlation of these data is analyzed, so as to more accurately predict the living habit of user, help to use
Family has managed home environment and has configured, and system output refers to the control instruction of the controlled household electrical appliances of ANFIS system outputs, target in figure
Value refers to family's electricity condition needed for actual user, and it is the inside of ANFIS systems to judge whether system output is consistent with desired value
Program obtains a stable user behavior custom prediction model, the custom prediction of stable user behavior by constantly comparing
After model is formed, the controlled household electrical appliances of control instruction control are being sent without user, are being sent automatically by user behavior custom prediction model
Control instruction.
Fig. 4 is the layer of structure figure of ANFIS systems of the present invention, and one is divided into 5 layers；ANFIS algorithms of the present invention are
The structure of the ANFIS of one typical multiple input and an output based on T-S patterns paste neural inference system；
First layer：This layer is conditional parameter；Each node i of this layer is an adaptive session for having node function
The node function of point, this layer is the membership function of fuzzy set, by it is determined that input x (or y) meets the journey of domain A (or B)
Degree.The membership function of domain A (or B) includes Gaussian function, trigonometric function and bell function etc., can be that any one is suitable
Parametrization membership function.What the present invention selected is Gaussian function；
Wherein, O1For the output of first layer,Respectively input the degree of membership of x (or y) each node i
Function；
The second layer：This layer is excitation density layer；Each node in this layer is fixed, its output is all
The algebraic product of input signal, formula are as follows：
Wherein, O2For the output of the second layer；
Third layer：This layer is that excitation density normalizes layer；Each node of this layer is also fixed, can not be trained；It indicates
Be that the excitation density of every fuzzy rule accounts for the percentage of strictly all rules excitation density, formula is as follows：
Wherein, O3For the output of third layer；wiIndicate the excitation density of each node i,It is fuzzy for each node i
The excitation density of rule accounts for the percentage of strictly all rules excitation density；
4th layer：This layer is consequent parameter layer；This layer is the adaptive node for having node function；Its node function includes
The parameter of conclusion part, and influence of the output to actual result of every rule can be calculated, formula is as follows：
Wherein, O4For the 4th layer of output；pi, qi, riFor the consequent parameter of this layer；
Layer 5：This layer is analytic fuzzy layer；The single node of this layer is the cumulative of all signals transmitted, and the layer is same
It is also stationary nodes, can not trains, formula is as follows：
Wherein, O5For the output of layer 5；f1, f2For node 1,2 corresponding function of node, generally linear function.
The parameter that Adaptive Neuro-fuzzy Inference can learn is divided into two parts：The premise parameter of first layer and the 4th layer
Consequent parameter.The form of premise parameter is related to the membership function of selection, the form of consequent parameter and of fuzzy rule
Number is related with the function selected.Learn consequent parameter first, in each iteration, input signal travels to forward along network
The 4th layer of network, fixed premise parameter is constant at this time, uses least square calligraphy learning consequent parameter；Then learn premise parameter, believe
Number continue to propagate to layer 5, be exported, keeps consequent parameter constant at this time, learnt using back propagation.This is learned
Habit process can iteration it is multiple.
Have the things of many strong correlations in real world, for example, the intensity of illumination L of light bulb and the power P of light bulb it is in direct ratio
Relationship, by taking the intensity of illumination in smart home as an example, he depend on the above-mentioned specific time, room number, what day, with the time,
Season, month relationship have no very big association；Common Nonlinear Mapping system, will when handling this type inputoutput data
All input variables disposably fully enter, and there is no distinguishing to contact power between input variable and result, result in and are
A series of problems that system input dimension is excessively high and causes, such as the training time is long, computation rule is excessive.
It illustrates：The concrete operating principle of the present invention is introduced with ANFIS systems the control of intensity of illumination below：
In data preprocessing module the input of single input ANFIS algorithms successively from the time, season, month, week, it is specific when
Between, input of one of factor as single input ANFIS algorithms is extracted in room number, device number, intensity of illumination；Data are located in advance
The output for managing single input ANFIS algorithms in module is the on off state of intelligent electric lamp；Time, season, month, week, it is specific when
Between, data by system time record can obtain, the number that room number, device number are distributed by ZigBee-network, intensity of illumination is by illumination
Sensor can measure；By the experience of life it is found that there are certainty in the intensity of light and the animation of people and specific time in family
Contact, such as in family nobody when or daytime sunlight intensity it is larger when without opening intelligent electric lamp；Therefore, intelligent electric lamp
On off state will have relationship with what day, specific time, intensity of illumination, unrelated with the factors such as time, season, month；First will
The input of single input ANFIS algorithms is combined with system output successively, forms multiple two-dimensional arrays, and such as (time, intelligent electric lamp are opened
Off status), (specific time, the on off state of intelligent electric lamp), using this two-dimensional array as the new input of single input ANFIS algorithms
Output, i.e. first layer；Single input ANFIS algorithmic systems determine membership function and fuzzy rule according to input and output, are then
The second layer of system is according to formula：Operation is made, third layer is transferred data to,
The excitation density for calculating every fuzzy rule accounts for the percentage of strictly all rules excitation density：Data
Consequent parameter is estimated using least squares estimate after being transferred to the 4th layer, signal continues forward direction and is sent to the 5th layer of output layer；So
Obtained error signal is completed into single single input ANFIS along channel backpropagation using BP algorithm regularization condition parameter afterwards
The primary study of algorithmic system obtains the minimal error e of single input ANFIS algorithmic systems after successive ignitioni；When all new
The single input ANFIS algorithms of binary array composition total i error parameter is obtained by training, pass through determination data dependence
Algorithm rejects the larger parameter of error, retains the ANFIS systems that parameter current is accustomed to as a user behavior in prediction model,
Complete the determination of true data dependence, while obtaining a trained learning-oriented user behavior custom prediction model, due to
The output of family behavioural habits prediction model is the on off state of intelligent electric lamp, so user behavior custom prediction model passes through instruction
Form switch state is communicated to intelligent electric lamp by ZigBee-network, realize the operation of control intelligent electric lamp.
The specific implementation mode of the present invention is explained in detail above in association with attached drawing, but the present invention is not limited to above-mentioned
Embodiment within the knowledge of a person skilled in the art can also be before not departing from present inventive concept
Put that various changes can be made.
Claims (9)
1. a kind of intelligent domestic system based on ANFIS algorithms, it is characterised in that：Including node device module, network management mould
Block, intelligent gateway, user management module；
Node device module is connect with network management module, and network management module is connect with intelligent gateway, and intelligent gateway is managed with family
Manage module connection；
User management module sends the control instruction that control is controlled household electrical appliances for user；
Node device module includes sensor device, controlled household electrical appliances；
Network management module is used to, by sensor and controlled household electrical appliances wireless networking, obtain room number, the device number for being controlled household electrical appliances, and
The environmental information that the room number, device number and sensor device of controlled household electrical appliances detect is sent to intelligent gateway, is used for simultaneously
The control instruction that intelligent gateway is sent is sent to controlled household electrical appliances；
Intelligent gateway includes controller, toy data base, data preprocessing module and user behavior custom prediction model, small data
Connect respectively with and with controller, data preprocessing module according to library, controller simultaneously with user management module, network management module
Connection；Controller is sent respectively to toy data base, network management module for user management module to be sent control instruction, and
Room number, the device number of the controlled household electrical appliances that network management module is sent are sent to toy data base, toy data base is used for
The time data of each controlled household electrical appliances is stored, and for receiving and storing control instruction, the sensor dress that user management module is sent
It sets the information sent, the room number and device number of the controlled household electrical appliances that network management module is sent, and the information of storage is sent
To data preprocessing module, data preprocessing module is used to receive the information that toy data base is sent and by determining that data are related
Property algorithm filter out the data larger with controlled household electrical appliances working condition correlation, be then passed to user behavior custom prediction mould
Type；User behavior custom prediction model is used to go out by ANFIS network analyses the behavioural habits of user, and it is controlled then to form control
The control instruction of household electrical appliances, and the control instruction of formation is sent to controlled household electrical appliances by network management module.
2. a kind of intelligent domestic system based on ANFIS algorithms according to claim 1, it is characterised in that：It is described really
The algorithm for determining data dependence includes single input ANFIS algorithms.
3. a kind of intelligent domestic system based on ANFIS algorithms according to claim 1, it is characterised in that：The biography
Sensor arrangement includes Temperature Humidity Sensor, optical sensor, and controlled household electrical appliances include air-conditioning, humidifier, intelligent electric lamp, smart window
Curtain.
4. a kind of intelligent domestic system based on ANFIS algorithms according to claim 1, it is characterised in that：The net
Network management module uses ZigBee wireless networkings, the ZigBee models CC2530 of use.
5. a kind of intelligent domestic system based on ANFIS algorithms according to claim 1, it is characterised in that：The use
Family management module is the ends PC or cell phone application end or Web page, and toy data base uses SQLite3.
6. a kind of intelligent domestic system based on ANFIS algorithms according to claim 1, it is characterised in that：The quilt
The time data of control household electrical appliances includes time, season, month, week, the specific time that controlled household electrical appliances use, user management module hair
The control instruction come refers to the on off state of controlled household electrical appliances.
7. a kind of intelligent domestic system based on ANFIS algorithms according to claim 6, it is characterised in that：The year
Part is the time for generating current data, is preserved with int types；
Season is to distinguish the on off state of the controlled household electrical appliances of Various Seasonal in 1 year, and numberical range is the int integers of 1-4；
It is to discriminate between the slight change that appliance switch state is controlled in the same season month, numberical range is the int types of 1-12；
Week is to distinguish the specific works state for time for being controlled household electrical appliances in a calendar month, and numberical range is the int of 1-7
Type；
The specific time is the specific works state for time that household electrical appliances are controlled in 24 hours one day, and time interval is every 10 minutes records one
Secondary, numerical value saves as int types；
Room number is to discriminate between the region of controlled household electrical appliances installation, and numberical range is the sum of the rooms 1- number；
Device number is to discriminate between the unique mark of different controlled household electrical appliances, and numberical range is the sum of 1- number of devices；
Equipment state is to show the working condition of current controlled household electrical appliances, and 0 expression current device is closed, and 1 indicates current
Equipment is in running order.
8. a kind of working method of the intelligent domestic system based on ANFIS algorithms, it is characterised in that：Include the following steps：
step1:Control command is sent to the controller in intelligent gateway by user by user management module, and controller will control
Instruction is sent to toy data base storage, while it is in the block that control instruction by network management module is sent to node device mould
Controlled household electrical appliances；
step2:Node device module acquires indoor environment Data Concurrent by sensor device and gives network management module；
step3：Network management module realizes the networking of sensor device and controlled household electrical appliances in family by ZigBee wireless,
Room number, the device number for being controlled household electrical appliances are obtained, and the room number, device number and sensor device of the controlled household electrical appliances of acquisition are examined
The environmental information measured is sent to the controller in intelligent gateway；
step4：The information that controller receiving network managing module in intelligent gateway is sent, and the information of reception is sent to small
Type database, toy data base itself are stored with the temporal information of controlled household electrical appliances, and toy data base believes the time of controlled household electrical appliances
Control instruction, the environmental information of sensor device for the user management module that breath, room number, device number and controller are sent are sent
To data preprocessing module, data preprocessing module by the information received by determine data dependence algorithm filter out with
The larger data of controlled household electrical appliances working condition correlation are then passed to user behavior custom prediction model；
step5：User behavior custom prediction model goes out the behavioural habits of user by ANFIS network analyses, then forms control
The control instruction of controlled household electrical appliances, and control instruction is automatically transmitted to controlled household electrical appliances by network management module.
9. a kind of working method of intelligent domestic system based on ANFIS algorithms according to claim 8, feature exist
In：In the step4, determine that the algorithm of data dependence includes the following steps：
step4.1：Build single input ANFIS：Input variable is combined with output variable respectively, obtains two tuple (x1, y), (x2,
Y) ..., (xn, y), two tuples are inputted into single input ANFIS, training network is learnt, and prediction result and energy are obtained
Obtain corresponding prediction error e1, e2..., en,
Wherein, xiI-th of input vector x is indicated respectively；Y is system output, eiFor i-th factor predicted value of input vector x with
The prediction error of actual value, i=1,2 ..., n；
step4.2：The percentage that each prediction error accounts for overall error is calculated, formula is as follows：
Wherein,The percentage of overall error is accounted for for each prediction error；
step4.3：Given threshold, and each of will calculate percentage and threshold value comparison that prediction error accounts for overall error, house
Prediction error is gone to account for input variable of the percentage more than given threshold of overall error；
step4.4：It is accustomed to the input of prediction model using the input variable preserved as user behavior.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810599563.5A CN108536030A (en) | 2018-06-12 | 2018-06-12 | A kind of intelligent domestic system and its working method based on ANFIS algorithms |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810599563.5A CN108536030A (en) | 2018-06-12 | 2018-06-12 | A kind of intelligent domestic system and its working method based on ANFIS algorithms |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108536030A true CN108536030A (en) | 2018-09-14 |
Family
ID=63470844
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810599563.5A Pending CN108536030A (en) | 2018-06-12 | 2018-06-12 | A kind of intelligent domestic system and its working method based on ANFIS algorithms |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108536030A (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109407641A (en) * | 2018-12-26 | 2019-03-01 | 北京理工华汇智能科技有限公司 | Intelligentized Furniture control device |
| CN109613902A (en) * | 2019-02-26 | 2019-04-12 | 网宿科技股份有限公司 | Smart home system and the method and Cloud Server for controlling smart home device |
| CN109634129A (en) * | 2018-11-02 | 2019-04-16 | 深圳慧安康科技有限公司 | Implementation method, system and the device actively shown loving care for |
| CN109669361A (en) * | 2018-12-26 | 2019-04-23 | 北京理工华汇智能科技有限公司 | Intelligentized Furniture system control method and device |
| CN109828474A (en) * | 2019-01-15 | 2019-05-31 | 深圳旦倍科技有限公司 | Cloud intelligent environment management method and system based on big data |
| CN109976167A (en) * | 2017-12-27 | 2019-07-05 | 丰田自动车株式会社 | Transportation system, information processing unit and information processing method |
| WO2021000790A1 (en) * | 2019-07-02 | 2021-01-07 | 珠海格力电器股份有限公司 | Smart home control method and smart home control apparatus |
| WO2021169752A1 (en) * | 2020-02-25 | 2021-09-02 | 青岛海尔洗衣机有限公司 | Edge computing device management method and apparatus, devices, and system |
| TWI763990B (en) * | 2019-04-22 | 2022-05-11 | 第一商業銀行股份有限公司 | Appraisal method and system of buildings based on urban and rural attributes |
| CN115801840A (en) * | 2022-12-05 | 2023-03-14 | 温州达派智能科技有限公司 | Big data detection system |
| CN117348434A (en) * | 2023-11-16 | 2024-01-05 | 佛山市康利家具有限公司 | Intelligent home management system based on user living habit |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104007741A (en) * | 2014-06-12 | 2014-08-27 | 深圳市智能帮科技有限公司 | Plug and play type intelligent housing system |
| US20150057808A1 (en) * | 2008-09-11 | 2015-02-26 | Washington State University, Office of Commercialization | Systems and Methods for Adaptive Smart Environment Automation |
| CN104486416A (en) * | 2014-12-16 | 2015-04-01 | 三星电子（中国）研发中心 | Comprehensive utilizing system and method of intelligent home service rule |
| CN105607508A (en) * | 2016-03-24 | 2016-05-25 | 重庆邮电大学 | Smart home device control method and system based on user behavior analysis |
| CN105955221A (en) * | 2016-06-21 | 2016-09-21 | 北京百度网讯科技有限公司 | Electric appliance equipment control method and apparatus |
| CN106292585A (en) * | 2016-08-16 | 2017-01-04 | 上海移为通信技术股份有限公司 | Smart Home poll networking system and method based on bluetooth group technology |
-
2018
- 2018-06-12 CN CN201810599563.5A patent/CN108536030A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150057808A1 (en) * | 2008-09-11 | 2015-02-26 | Washington State University, Office of Commercialization | Systems and Methods for Adaptive Smart Environment Automation |
| CN104007741A (en) * | 2014-06-12 | 2014-08-27 | 深圳市智能帮科技有限公司 | Plug and play type intelligent housing system |
| CN104486416A (en) * | 2014-12-16 | 2015-04-01 | 三星电子（中国）研发中心 | Comprehensive utilizing system and method of intelligent home service rule |
| CN105607508A (en) * | 2016-03-24 | 2016-05-25 | 重庆邮电大学 | Smart home device control method and system based on user behavior analysis |
| CN105955221A (en) * | 2016-06-21 | 2016-09-21 | 北京百度网讯科技有限公司 | Electric appliance equipment control method and apparatus |
| CN106292585A (en) * | 2016-08-16 | 2017-01-04 | 上海移为通信技术股份有限公司 | Smart Home poll networking system and method based on bluetooth group technology |
Non-Patent Citations (1)
| Title |
|---|
| WANGLEI等: "Intelligent Control in Smart Home based on Adaptive Neuro Fuzzy Inference System", 《2015 CHINESE AUTOMATION CONGRESS (CAC)》 * |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109976167A (en) * | 2017-12-27 | 2019-07-05 | 丰田自动车株式会社 | Transportation system, information processing unit and information processing method |
| CN109634129B (en) * | 2018-11-02 | 2022-07-01 | 深圳慧安康科技有限公司 | Method, system and device for realizing active care |
| CN109634129A (en) * | 2018-11-02 | 2019-04-16 | 深圳慧安康科技有限公司 | Implementation method, system and the device actively shown loving care for |
| CN109669361A (en) * | 2018-12-26 | 2019-04-23 | 北京理工华汇智能科技有限公司 | Intelligentized Furniture system control method and device |
| CN109407641A (en) * | 2018-12-26 | 2019-03-01 | 北京理工华汇智能科技有限公司 | Intelligentized Furniture control device |
| CN109828474A (en) * | 2019-01-15 | 2019-05-31 | 深圳旦倍科技有限公司 | Cloud intelligent environment management method and system based on big data |
| CN109613902A (en) * | 2019-02-26 | 2019-04-12 | 网宿科技股份有限公司 | Smart home system and the method and Cloud Server for controlling smart home device |
| TWI763990B (en) * | 2019-04-22 | 2022-05-11 | 第一商業銀行股份有限公司 | Appraisal method and system of buildings based on urban and rural attributes |
| WO2021000790A1 (en) * | 2019-07-02 | 2021-01-07 | 珠海格力电器股份有限公司 | Smart home control method and smart home control apparatus |
| WO2021169752A1 (en) * | 2020-02-25 | 2021-09-02 | 青岛海尔洗衣机有限公司 | Edge computing device management method and apparatus, devices, and system |
| CN115801840A (en) * | 2022-12-05 | 2023-03-14 | 温州达派智能科技有限公司 | Big data detection system |
| CN115801840B (en) * | 2022-12-05 | 2023-12-15 | 北京金安道大数据科技有限公司 | Big data detection system |
| CN117348434A (en) * | 2023-11-16 | 2024-01-05 | 佛山市康利家具有限公司 | Intelligent home management system based on user living habit |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108536030A (en) | A kind of intelligent domestic system and its working method based on ANFIS algorithms | |
| Han et al. | A review of reinforcement learning methodologies for controlling occupant comfort in buildings | |
| Amasyali et al. | Machine learning for occupant-behavior-sensitive cooling energy consumption prediction in office buildings | |
| CN107120782B (en) | A kind of HVAC system control method based on multi-user's hot comfort data | |
| KR101875488B1 (en) | Method and system for automatic controlling of air conditioner by using an artificial intelligence | |
| Jin et al. | A data-driven model predictive control for lighting system based on historical occupancy in an office building: Methodology development | |
| KR101875489B1 (en) | Method and system for automatic controlling of air conditioner by using an artificial intelligence | |
| CN110377084B (en) | Building indoor environment regulation and control method based on intelligent control strategy | |
| CN104133427A (en) | Intelligent household control method and system | |
| CN105320184A (en) | Intelligent monitoring system of indoor environment of building | |
| CN113902582A (en) | Building comprehensive energy load prediction method and system | |
| CN110471301A (en) | A kind of smart home service recommendation system and method based on user behavior | |
| Rambabu et al. | Prediction and analysis of household energy consumption by machine learning algorithms in energy management | |
| Sun et al. | Energy consumption optimization of building air conditioning system via combining the parallel temporal convolutional neural network and adaptive opposition-learning chimp algorithm | |
| Deng et al. | Toward smart multizone HVAC control by combining context-aware system and deep reinforcement learning | |
| Fouladfar et al. | Adaptive thermal load prediction in residential buildings using artificial neural networks | |
| Walek et al. | Adaptive fuzzy control of thermal comfort in smart houses | |
| Fu et al. | A Sarsa-based adaptive controller for building energy conservation | |
| Xiang et al. | Prediction model of household appliance energy consumption based on machine learning | |
| Yin et al. | Personalized ambience: an integration of learning model and intelligent lighting control | |
| Daum | On the Adaptation of Building Controls to the Envelope and the Occupants | |
| Qiao | Intelligent building with multi-energy system planning method considering energy supply reliability | |
| Omarov | Development of fuzzy based smart building energy and comfort management system | |
| Zhang | Data-driven whole building energy forecasting model for data predictive control | |
| Wang et al. | Study on factors correlation of personal lighting comfort model in cyber-physical human centric 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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180914 |