CN105910225A - Air conditioner load control system and method based on personnel information detection - Google Patents
Air conditioner load control system and method based on personnel information detection Download PDFInfo
- Publication number
- CN105910225A CN105910225A CN201610244200.0A CN201610244200A CN105910225A CN 105910225 A CN105910225 A CN 105910225A CN 201610244200 A CN201610244200 A CN 201610244200A CN 105910225 A CN105910225 A CN 105910225A
- Authority
- CN
- China
- Prior art keywords
- air conditioner
- conditioner load
- personal information
- personnel
- air
- 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
Classifications
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
- F24F11/00—Control or safety arrangements
- F24F11/30—Control or safety arrangements for purposes related to the operation of the system, e.g. for safety or monitoring
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
- F24F11/00—Control or safety arrangements
- F24F11/62—Control or safety arrangements characterised by the type of control or by internal processing, e.g. using fuzzy logic, adaptive control or estimation of values
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Mathematical Physics (AREA)
- General Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- Evolutionary Computation (AREA)
- Data Mining & Analysis (AREA)
- Artificial Intelligence (AREA)
- Computational Linguistics (AREA)
- Human Resources & Organizations (AREA)
- Computing Systems (AREA)
- General Health & Medical Sciences (AREA)
- Economics (AREA)
- Biomedical Technology (AREA)
- Chemical & Material Sciences (AREA)
- Combustion & Propulsion (AREA)
- Mechanical Engineering (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Operations Research (AREA)
- Signal Processing (AREA)
- Fuzzy Systems (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Quality & Reliability (AREA)
- Marketing (AREA)
- Entrepreneurship & Innovation (AREA)
- Game Theory and Decision Science (AREA)
- Development Economics (AREA)
- Air Conditioning Control Device (AREA)
Abstract
The invention discloses an air conditioner load control system based on personnel information detection. The air conditioner load control system comprises a BP neural network module used for completing training and predicting the air conditioner load quantity and further comprises a personnel information detection module used for detecting personnel information within the air conditioner working range. The personnel information comprises the personnel types and the corresponding quantities. The BP neural network module is provided with a corresponding relation between an air conditioner load and the personnel information. The invention further discloses an air conditioner load control method based on personnel information detection. According to the air conditioner load control system and method based on personnel information detection, compared with other personnel number acquisition technologies used for air conditioner load predicting, the personnel load requirement difference can be better analyzed, and the air conditioner load control system and method have very important significance for precise load management and control, internal personnel thermal comfort strengthening and comfortable artificial environment creating.
Description
Technical field
The present invention relates to management and the control field of air conditioner load, particularly to one based on personal information
The air conditioner load control system of detection and method.
Background technology
The energy and environmental problem are the big hot issues of current two, and wherein building energy consumption account for total energy
15~more than 60% consumed, are the important component parts realizing the strategy of sustainable development.To building energy
Consumption carries out the management of effective science and is possible not only to improve the utilization ratio of the energy, it is possible to reduce titanium dioxide
The discharge of carbon.Artificial intelligence is the new lover of 21 century development in science and technology, and " degree of depth study " is dirty when being
The artificial intelligence technology of row, as the frontier of research machine learning, it is intended to by imitating human brain god
Analyze through network, simulation human brain study mechanism, process data.Artificial intelligence is not only used in
In go Great War, more can predict at building energy consumption, management aspect plays an important role.
In building energy consumption, the power consumption of the public building such as market is the hugest, with 10, Beijing megastore
As a example by, year power consumption about 1,000,000,000 degree, wherein air-conditioning energy consumption just account for about 50%, therefore business
The energy-saving potential of field air conditioner load is the hugest.And present stage market is meet internal staff comfortable
Property requirement, often with maximum number that market is likely to be breached on the same day as total number of persons, calculate prediction maximum
Air conditioner load amount, and do not consider the factors such as concrete personnel characteristics, active state, comfortable in market
Property reduce while also easily cause the waste of energy.Thus real-time, personnel's letter accurately
Breath statistical system brings huge facility can to the air conditioner load management in market, saves the same of the energy
Time also can strengthen the comfortableness of client, bring huge economic benefit to market.Personal information is added up
With combining of Air-conditioning Load Prediction, building energy consumption management illustrates fabulous potentiality, for
The building of development green health is significant.
Thermal comfort refers to personnel's state of consciousness to environment representation satisfaction above, and human body is the warm of self
Equilibrium condition and the environmental aspect felt integrate the sensation judging whether to reach comfortable, including raw
Reason and two aspects of psychology.The influence factor of Thermal comfort both included temperature, humidity, vertical temperature-difference,
The environmental factorss such as blowing amount, season, also include the personal informations such as such as age, sex, activity intensity.
BP neutral net is the processing unit neuron most basic by being similar to human brain in a large number, extensively
General being coupled to each other and the intelligent network system of non-linear complexity that constitutes.BP neutral net is as one
Effective building air-conditioning load prediction mode, the most gradually receives field of heating ventilation air conditioning professional people
The favor of scholar.The such as patent documentation of application publication number CN 104008427A disclose a kind of based on
The Forecasting Methodology of the central air-conditioning refrigeration duty of BP neutral net, comprises the following steps: step 1, choosing
Select and affect the factor of School Buildings refrigeration duty as nerve network input parameter;Step 2, to building cold
Load prediction sample data carries out arranging and pretreatment;Step 3, the level of design BP neutral net
Structure, determines the implicit number of plies;Step 4, operation BP neural metwork training, until reversely convergence,
Terminate study, export predictive value.Said method has accuracy and high reliability.
In existing BP neural network prediction market, air conditioner load does not the most consider the impact of human factor
Or owing to the problems such as technology closure can not be predicted in real time, therefore exist adaptability the most by force with
And the shortcoming such as temporal hysteresis quality.
Summary of the invention
The invention discloses a kind of air conditioner load control system based on personal information detection, control ginseng
Number adds the personal information parameter of hommization, and can detect in real time, to reach to carry
High-comfort and energy-conservation effect.
A kind of air conditioner load control system based on personal information detection, including completing training in advance
Survey the BP neural network module of air conditioner load amount, also include: personal information detection module, be used for examining
Surveying the personal information in the range of air-conditioning work, described personal information is personnel's type and corresponding number
Amount;The corresponding relation that described BP neural network module is provided with between air conditioner load and personal information.
The mode obtaining personal information is a lot, it is preferred that described personal information detection module includes arranging
Image unit in the range of air-conditioning work and receive the image information of image unit and export personnel
Type and Opencv function library static state personnel's information calculating unit of corresponding quantity.
Present invention utilizes Opencv static state personnel's information detection technology, personal information in building is entered
Row takes into full account and detection process reaches the requirement of " in real time ", image unit and Opencv function library
Static personnel's information calculating unit is connected with BP neural network module, more simultaneously by sensor after connecting
Module (being used for detecting other environmental variables), automaticdata capture input module and BP neutral net
Module connects, and is finally connected with air-conditioning equipment control module by BP neural network module, all connections
Process is reached by transport module, and the BP neural network module real-time estimate output complete by training is built
Build internal air conditioner load and control, and accomplish automatic prediction by hardware such as PCs and regulate and control.
Sensor assembly includes controlling single-chip microcomputer, the reality that the digital independent such as temperature, humidity forwards with bluetooth
The bluetooth serial ports of existing intermodule communication function, the temperature sensor of acquisition temperature information, acquisition humidity letter
The humidity sensor of breath, the keil programming of single-chip microcomputer comprises temperature reading, humidity reads, serial ports.
Image unit inputs real-time video to Opencv function library static state personnel's information calculating unit, from
Dynamic data capture module automatically captures Essential Environment amount in official's external server and inputs BP nerve
Mixed-media network modules mixed-media, sensor inputs the influence factors such as building interior humiture to BP neural network module,
Opencv function library static state personnel's information calculating unit inputs real-time personnel to BP neural network module
Information, BP neural network module inputs real-time air-conditioning prediction load, air-conditioning control to airconditioning control module
Molding block controls air-conditioning equipment, and each equipment is connected by transport module, and sensor can be by bluetooth etc.
Communication apparatus inputs data to BP neural network module, and all processes are all by hardware such as PCs certainly
Dynamic regulation and control.
A kind of air conditioner load control method based on personal information detection, uses above-mentioned based on personnel
The air conditioner load control system of infomation detection, comprises the following steps:
(1) by right between collector's information data and air conditioner load and personal information in advance
Should be related to and BP neural network module is trained;
(2) personal information detection module detects the personnel's type in the range of air-conditioning work and right in real time
The quantity answered, and personal information result is sent to the BP neutral net that in step (1), training is complete
Module;
(3) BP neural network module is according to receiving real-time personal information prediction current time air-conditioning man
Air-conditioning is also controlled by air conditioner load required in the range of work;
(4) circulation step (2) and (3).
BP neutral net is the processing unit neuron most basic by being similar to human brain in a large number, extensively
General being coupled to each other and the intelligent network system of non-linear complexity that constitutes.The basic think of of BP learning algorithm
Think that teacher of the employing learns exactly, on the basis of given substantial amounts of input/output signal, set up system
Data are carried out large-scale parallel processing by non-linear input/output model, by network output error
Back propagation, adjust and revise the connection weights of network, make error minimize, its learning process
Including forward calculation and error back propagation.
Obtaining air conditioner load and by data acquisition on the spot in BP neural metwork training module is
Many groups corresponding relation between row influence factor, utilize neutral net connection entropy and its self study,
The function such as self adaptation, nonlinear mapping, obtains above-mentioned non-linear complex relationship between the two.BP
Neural metwork training module can also utilize history air conditioner load database data be analyzed instruction
Practice, efficiently utilize big data air conditioner load to be predicted, monitor and adjusts.
In the present invention, the data of BP neural metwork training module include personal information, described personal information
For personnel's type and corresponding quantity, in order to improve efficiency and accuracy, the present invention uses based on
The personal information detecting system of Opencv function library obtains personal information, and this system is that shooting is single
Unit obtains sequence of video images in real time and automatically analyzes.One video flowing is by the image combination of multiframe
Becoming, finder from image, for every two field picture, this is a static target, makes process obtain
To simplify, it is programmed by Opencv function library, video is detected, available taken
All kinds of personal informations in this two field picture corresponding to quarter, including child's number (≤14 years old), person between twenty and fifty
Number (14~60 years old), old people's number (>=60 years old), male's number, women number, it is in fortune
The personal informations such as the number of dynamic state, the number remained static, can adjust parameter as required
Scope, different types of personnel are different to the requirement of temperature, and the inventive method can be according to dissimilar
The temperature requirements of personnel air-conditioning is carried out overall regulation and control, strengthen personnel's thermal comfort.
Human body carries out the energy of activity from the heat produced in vivo in metabolic processes, therefore,
The metabolism rate of human body directly affects the heat exchange of human body and surrounding.The relative metabolic rate of human body
Can affected by various factors, as musculation intensity, ambient temperature height, sex, the age,
Race and living habit etc..The metabolism rate of the people of different age group is different, such as between twenty and fifty metabolic rate
By force, and the difference of activity intensity also has the biggest impact to metabolism, such as the crowd of aggravating activities
Metabolic rate is high, is more biased towards in the relatively low environment of temperature.When building interior person between twenty and fifty's number or be in motion
The number of state increases, and total metabolism strengthens, and needed for building interior, refrigeration duty increases the most therewith, need to fit
When the supply increasing air conditioner load.
Different sexes crowd also has different requirements to ambient temperature, such as women relative to male the most more
Liking the clothes that comparison is frivolous, the thermal resistance of medicated clothing is little, dispels the heat more, and therefore women is more biased towards in temperature
Higher environment.When building interior women number increases relatively, need to suitably reduce the confession of air conditioner load
Give to meet comfortableness requirement.
Building interior people is known precisely by personal information detecting system based on Opencv function library
The specifying information of member, concrete number such as sex, age with whether be kept in motion and corresponding
Amount, and the most suitably adjusts air conditioner load, to strengthen internal staff thermal comfort and comfortable manually
The construction of environment is significant.
The data trained in the present invention also include real-time air conditioner load amount and real-time environmental variable
The air conditioner load amount of data and the same period in former years and previous environmental variable data.Described environmental variable bag
Include Discussion on architecture humiture, architectural exterior-protecting construction parameter, build remaining series such as equipment cooling, illumination
Affect the variable of building air conditioning load.
Preferably, whether described personnel's type includes: sex, age and be kept in motion
At least one.Above-mentioned all types of can all be set, it is also possible to select according to applicable situation
Select.
Preferably, in step (2), personal information detection module detects in real time in air-conditioning work scope
The concrete grammar of interior personnel's type and corresponding quantity comprises the following steps:
2-1, image unit gather the image in the range of air-conditioning work;
2-2, employing Opencv function library static state personnel's information calculating unit receive the image of image unit
Information also exports personnel's type.
In order to reduce amount of calculation, improve computational efficiency, it is preferred that in step (2), every one
Time segment t detects real time personnel information a in this momentij, take n time period t and predict averagely
Personal informationAnd result is sent to BP neutral net.
In order to improve the control effect of this method, it is considered to the shadow of other environmental factorss such as indoor and outdoor humiture
Ring, obtain air conditioner load by the prediction of BP nerual network technique, and utilize PC automatically to control
System, adds data and temporal accuracy, to reach energy-conservation effect, it is preferred that step (1)
In, gather environmental variable data right according between air conditioner load and environmental variable the most in advance
Should be related to and BP neural network module is trained;
In step (2), instruction in detecting environmental variable the most in real time and sending result to step (1)
Practice complete BP neural network module;
In step (3), BP neural network module receives real-time environmental variable simultaneously and combines personnel
Air-conditioning is also controlled by air conditioner load required in the range of information prediction current time air-conditioning work.
Preferably, environmental variable includes inside and outside humiture, architectural exterior-protecting construction parameter, builds remaining and set
At least one in standby heat radiation and illumination.
In order to improve the accuracy of control, during the t obtained eventually through the output layer of BP neutral net
Carve refrigeration duty optimal solution, update a period of time under central air-conditioning according to each indoor environment variable perturbations trend simultaneously
The air conditioner load carved, it is preferred that in step (4), it was predicted that | r 'it| with | rit| difference, when t+1
The prediction air conditioner load value carved adds described difference, | r 'it| for air-conditioning actual load amount, | rit| pre-for air-conditioning
Survey loading.
The complete BP neutral net of training in the present invention is with all kinds of personal informations, t Indoor Temperature
Humidity, t outdoor temperature humidity, ventilation rate, t-1 moment refrigeration duty etc. are input layer parameter, middle
Hidden layer each input value is weighted and threshold value judgement, the output of hidden layer is carried out by output layer
Last processing, obtains t refrigeration duty optimal solution.
The present invention is by reading the information of monitor video, and personal informations all kinds of to building interior are carried out in real time
Ground reads, and other factor parameters utilizing measurement, acquiring, and carries out building air conditioning load more
Add and predict in real time, accurately and control, omnidistance by PC automatic Prediction, regulation and control, improve building
Energy consumption is reduced while artificial environment level of comfort.
The present invention also takes full advantage of existing equipment and information resources, is the existing degree of depth to information resources
Developing, system is simple and need not extra equipment, cost-effective.It addition, arrange with special
The methods such as passage are compared, and use monitoring to be not easy to affect the shopping desire of client, strengthen the recessiveness in market
Competitiveness.
Beneficial effects of the present invention:
The air conditioner load control system based on personal information detection of the present invention and method and other numbers
Acquiring technology is compared for Air-conditioning Load Prediction, can preferably analyze the workload demand difference of personnel,
The thermal comfort that with regulation and control, internal staff is strengthened for accurate load management and comfortable artificial environment
Construction have very important significance.
Accompanying drawing explanation
Fig. 1 is the wire frame structure of the air conditioner load control system based on personal information detection of the present invention
Figure.
Fig. 2 is the wire frame flow process of the air conditioner load control method based on personal information detection of the present invention
Figure.
Fig. 3 is the workflow diagram of the photographing module in apparatus of the present invention.
Fig. 4 is the sensor assembly workflow diagram in apparatus of the present invention.
Detailed description of the invention
As it is shown in figure 1, the air conditioner load control system based on personal information detection of the present embodiment includes:
Including completing to train for predicting the BP neural network module of air conditioner load amount, personal information detection mould
Block, for detection personal information in the range of air-conditioning work, described personal information be personnel's type and
Corresponding quantity;It is corresponding that described BP neural network module is provided with between air conditioner load with personal information
Relation, personal information detection module includes image unit and believes based on Opencv function library static state personnel
Breath computing unit, image unit be input to the monitor video obtained in real time to have set complete based on
In Opencv function library static state personnel's information calculating unit, image unit includes multiple photographic head, covers
Cover the work space (referring specifically to the scope that someone walks about) of whole air-conditioning.
Having used the identification module of HOG in the present embodiment, Opencv provides default models, pattern number
According to being that the sample INRIAPerson.tar training provided by existing document is obtained, the full name of HOG is
Histograms of Oriented Gradients, as the term suggests, histograms of oriented gradients, for target
A kind of describing mode.
The corresponding characteristic vector of each target, hog detection is 3781 dimensions, and hog is by a spy
Levy window (win) and be divided into a lot of blocks (block), be divided into again a lot thin in each block
Born of the same parents' unit (cell i.e. cell element), little characteristic vector corresponding for all of cell is strung by hog characteristic vector
Obtain the characteristic vector of a higher-dimension.Utilize hog+svm technology for detection pedestrian, final detection side
Method is most basic linear discriminant function, wx+b=0.Need to extract some training samples during training
This hog feature, its objective is w and b for obtaining detection.Mesh to be detected is extracted during detecting
Target hog feature x, is brought in equation and differentiates, can determine whether whether target is to need detection
Pedestrian.
By based on Opencv function library static state personnel's infomation detection, obtain monitoring in video and arbitrarily select
All kinds of personal informations in this frame still image are carved in timing, including child's number, between twenty and fifty number, old age
People's number, male's number, women number, the number being kept in motion, the people that remains static
Number etc..The ambient parameter that all kinds of personal informations obtained and monitoring and official obtain is inputted to having trained
In the BP neural network module finished.
BP network is a kind of multilayer feedforward type network, and typical network structure includes defeated people's layer, implies
Layer, output layer, if hidden layer can have dried layer.Need to instruct BP neutral net before input parameter
Practice.
The particular content set as:
The input layer of BP neutral net and hidden layer neuron use Sigmoid function, and output layer takes
Sigmoid function or linear function, use error-correction rule to learn, have parallel distributed structure.
BP learning algorithm uses teacher of the having to learn, and does not directly give the parsing relation of input and outlet chamber,
On the basis of given substantial amounts of input/output signal, set up the non-linear input/output model of system, right
Data carry out large-scale parallel processing, by the back propagation of network output error, adjust and revise
The connection weights of network, make error minimize, and learning process includes that forward calculation and error reversely pass
Broadcast.
Corresponding computing formula is as follows.
Hidden node is output as:
Output layer node is output as:
Error back propagation is calculated as:
Wherein xiRepresent input signal;yhRepresent the output of hidden node;zjRepresent the defeated of output node
Go out;ωihConnection weights for input node to hidden node;ωhjOutput node is arrived for hidden node
Connect weights;θhFor the threshold value of hidden node, γjFor the threshold value of output node, EkIt is that kth characterizes
The error of vector;yjkIt it is the expected value of jth output neuron;OjkIt it is jth output neuron
Actual value.
In the present embodiment by all kinds of personal informations, t indoor temperature and humidity, t outdoor temperature humidity,
Ventilation rate, t-1 moment refrigeration duty are input layer parameter, and each input value is added by middle hidden layer
Power and threshold value are adjudicated, and output layer carries out last processing to the output of hidden layer, obtain t cold negative
Lotus optimal solution.To given training sample, utilizing propagation formula, the direction reduced along error is continuous
Adjust network joint weights and threshold value, finally make the neural network learning of this prediction air conditioner load complete.
A large amount of air conditioner load numerical value and influence factor thereof is obtained by the data acquisition of a period of time
Between corresponding relation, or by calling history air conditioner load database data, be applied to BP neural
Network is trained, and obtains the BP nerve network system trained.
As in figure 2 it is shown, all kinds of personal informations that personal information detection module is obtained and monitoring and official
The ambient parameter obtained inputs to training in complete BP nerve network system, it was predicted that when obtaining current
Carve the concrete numerical value of air conditioner load needed for building interior, thus for the control of air conditioner load.
The input parameter of BP neutral net has indoor all kinds of personal information, including child's number p1(≤
14 years old), between twenty and fifty number p2(14~60 years old), old people's number p3(>=60 years old), male's number
p4, women number p5, number p that is kept in motion6, number p that remains static7, remaining
Personnel state information p8;Also have environmental variable (Temperature Humidity Sensor must build in Essential Environment amount,
Automatic Program captures the Essential Environment amounts such as the extraneous humiture of official's issue), including: outdoor temperature T0,
Indoor temperature Ti, outside humidity s0, indoor humidity si, room ventilation rate Qv, the load in a upper moment
Amount W-1;Also being output as loading W of air-conditioning, the computing formula of W is as follows:
W=f (p1,p2,p3,p4,p5,p6,p7,p8,T0,Ti,s0,si,Qv,W-1)
Above-mentioned formula is the neural metwork training function of load being best suitable for this market out, this control system
The main purpose of system is giving treatment in accordance with seasonal conditions and treatment in accordance with local conditions, can obtain optimal market load in real time,
Automatically controlled the loading of central air-conditioning by PC, reach energy-conservation effect, preferably analyze people
The workload demand difference of member, for accurate load management and regulation and control, the thermal comfort of enhancing internal staff
The construction of property and comfortable artificial environment has very important significance.
As it is shown on figure 3, the photographing module of the present embodiment uses based on H264 network monitoring camera head system
System, uses Leonardo da Vinci family chip DM8148 as the primary processor of system, and hardware platform mainly wraps
Include CCD camera, TVP5158 chip, DM8148 primary processor, LCD liquid crystal display screen, 1000M
Ethernet interface and solid-state SATA hard disc etc..By monitoring camera head module by video feed to client
End is to read personal information in real time.
CCD camera converts light signals into the signal of telecommunication and exports control centre, uses TVP5158
Video analog signal is converted to digital video signal by chip, in input primary processor, has sensitivity
High light high, anti-, distort the advantages such as little, volume is little, life-span length, anti-vibration.
DM8148 can directly carry solid-state SATA hard disc, have dual-port 1000M ethernet port,
And support the function such as decompression of H.264 form, thus select DM8148 as the main process of system
Device.
Use Leonardo da Vinci family chip DM8148 as the primary processor of system, obtain by calling driving
Take video data, send into LCD liquid crystal display screen and show and video original data is sent to Video coding
H.264, device encodes.Encoded good video data is stored in local solid-state SATA hard disc and RTP
Package utilizes 1000M Ethernet interface to realize network transmission.Terminal Server Client (PC) is by receiving
Video data stream, is decoded, plays, it is achieved real-time video monitoring.
As shown in Figure 4, the sensor assembly of the present embodiment includes Single-chip Controlling space, Bluetooth transmission
Assembly, sensor assembly.Sensor device is to utilize metal to have differential responses to reach to read to varying environment
Take the purpose of external environment parameters, bluetooth transmission means use wireless radio frequency transmission data reach by
Data are transferred to the function of processor.
Specifically include: Arduino UNO development board 1, HC06 bluetooth serial ports 2, DS18B20 temperature
Sensor (output valve temperature is-50~80 DEG C) 3 and DHT11 humidity sensors 4.
Arduino UNO development board is main control section, and single-chip microcomputer is IC chip, is to use
Very large scale integration technology is having the central processor CPU of data-handling capacity, storing at random
Device RAM, read only memory ROM, multiple I/O mouth and the interruption merit such as system, timer/counter
Can, the data such as this system utilizes the function of the similar microcomputer of single-chip microcomputer, it is achieved temperature, humidity
Read with control bluetooth module forwarding.
DS18B20 temperature sensor (output valve temperature is-50~80 DEG C) utilizes high-temperature coefficient crystal oscillator
Varying with temperature the characteristic that its oscillation rate substantially changes, obtain temperature information, encapsulation is in the circuit board
Can be with single chip communication.
DHT11 humidity sensor, stability is strong, can meet one the stable input of system needs
Demand, encapsulation gets final product and single chip communication in the circuit board.
HC06 bluetooth serial ports is connected by circuit with single-chip microcomputer, the data of caching single-chip microcomputer output port
And set up bluetooth with server and be connected, send data, it is achieved the communication function of intermodule.
Gathered in video input PC by monitoring camera head module, utilize set complete based on
Opencv function library static state personnel's information detecting system exports the concrete personnel of this moment building interior in real time
Information, including child's number, between twenty and fifty number, old people's number, male's number, women number,
The number being kept in motion, the number etc. remained static.By all kinds of personal informations sensor
Official's Essential Environment amount that the humiture parameter of module input, automaticdata crawl input module obtain is defeated
Enter to setting and training in complete BP neutral net, it was predicted that draw the optimal air conditioner load of current time
Amount.The real-time air-conditioning prediction loading automatically derived by PC inputs to airconditioning control module, control
Air-conditioning equipment processed runs.
Claims (9)
1. based on personal information detection an air conditioner load control system, including complete train for
The BP neural network module of prediction air conditioner load amount, it is characterised in that also include: personal information is examined
Surveying module, for detection personal information in the range of air-conditioning work, described personal information is personnel's class
Type and corresponding quantity;Described BP neural network module is provided with between air conditioner load and personal information
Corresponding relation.
2. the air conditioner load control system detected based on personal information as claimed in claim 1, its
Being characterised by, described personal information detection module includes being arranged in the image unit in the range of air-conditioning work
And receive the image information of image unit and export personnel's type and the Opencv letter of corresponding quantity
Number storehouse static state personnel's information calculating unit.
3. an air conditioner load control method based on personal information detection, it is characterised in that use
Air conditioner load control system based on personal information detection described in claim 1 or 2, including following
Step:
(1) by right between collector's information data and air conditioner load and personal information in advance
Should be related to and BP neural network module is trained;
(2) personal information detection module detects the personnel's type in the range of air-conditioning work and right in real time
The quantity answered, and personal information result is sent to the BP neutral net that in step (1), training is complete
Module;
(3) BP neural network module is according to receiving real-time personal information prediction current time air-conditioning man
Air-conditioning is also controlled by air conditioner load required in the range of work;
(4) circulation step (2) and (3).
4. the air conditioner load control method detected based on personal information as claimed in claim 3, its
Whether be characterised by, described personnel's type includes: sex, age and be kept in motion
At least one.
5. the air conditioner load control method detected based on personal information as claimed in claim 3, its
Being characterised by, in step (2), personal information detection module detects in real time in the range of air-conditioning work
Personnel's type and the concrete grammar of corresponding quantity comprise the following steps:
2-1, image unit gather the image in the range of air-conditioning work;
2-2, employing Opencv function library static state personnel's information calculating unit receive the image of image unit
Information also exports personnel's type.
6. the air conditioner load control method detected based on personal information as claimed in claim 3, its
It is characterised by, in step (2), detects the real time personnel information in this moment every time segment t
aij, take n time period t and predict average personal informationAnd send result to BP
Neutral net.
7. the air conditioner load control method detected based on personal information as claimed in claim 3, its
It is characterised by, in step (1), gathers environmental variable data the most in advance and according to air conditioner load
And BP neural network module is trained by the corresponding relation between environmental variable;
In step (2), instruction in detecting environmental variable the most in real time and sending result to step (1)
Practice complete BP neural network module;
In step (3), BP neural network module receives real-time environmental variable simultaneously and combines personnel
Air-conditioning is also controlled by air conditioner load required in the range of information prediction current time air-conditioning work.
8. the air conditioner load control method detected based on personal information as claimed in claim 7, its
Being characterised by, described environmental variable includes inside and outside humiture, architectural exterior-protecting construction parameter, building
At least one in remaining equipment cooling and illumination.
9. the air conditioner load control method detected based on personal information as claimed in claim 3, its
It is characterised by, in step (4), it was predicted that | r 'it| with | rit| difference, at the prediction air-conditioning in t+1 moment
Load value adds described difference, | r 'it| for air-conditioning actual load amount, | rit| predict loading for air-conditioning.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610244200.0A CN105910225A (en) | 2016-04-18 | 2016-04-18 | Air conditioner load control system and method based on personnel information detection |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610244200.0A CN105910225A (en) | 2016-04-18 | 2016-04-18 | Air conditioner load control system and method based on personnel information detection |
Publications (1)
Publication Number | Publication Date |
---|---|
CN105910225A true CN105910225A (en) | 2016-08-31 |
Family
ID=56747351
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610244200.0A Pending CN105910225A (en) | 2016-04-18 | 2016-04-18 | Air conditioner load control system and method based on personnel information detection |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN105910225A (en) |
Cited By (24)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106895564A (en) * | 2017-03-27 | 2017-06-27 | 中国科学院广州能源研究所 | A kind of station air conditioner control system and method |
CN106918119A (en) * | 2017-03-23 | 2017-07-04 | 上海巽晔计算机科技有限公司 | A kind of low central air conditioner system of operation risk |
CN107461875A (en) * | 2017-07-07 | 2017-12-12 | 珠海格力电器股份有限公司 | Control method and device of air conditioning system |
CN107883541A (en) * | 2017-10-24 | 2018-04-06 | 珠海格力电器股份有限公司 | Air conditioner control method and device |
CN108413588A (en) * | 2018-02-12 | 2018-08-17 | 北京工业大学 | A kind of personalized air-conditioner control system and method based on thermal imaging and BP neural network |
CN108647817A (en) * | 2018-05-07 | 2018-10-12 | 北京工业大学 | A kind of prediction technique and forecasting system of energy consumption load |
CN108664452A (en) * | 2018-05-07 | 2018-10-16 | 北京工业大学 | A kind of indoor occupant number determination method and determining system |
EP3379163A3 (en) * | 2016-12-19 | 2018-10-31 | Lg Electronics Inc. | Air conditioner and control method thereof |
CN109140665A (en) * | 2018-08-31 | 2019-01-04 | 珠海格力电器股份有限公司 | Method and device for controlling air conditioner and air conditioner device |
CN109164707A (en) * | 2018-09-28 | 2019-01-08 | 苏州市建筑科学研究院集团股份有限公司 | A kind of indoor environment negative-feedback regu- lation system based on artificial neural network algorithm |
CN110793162A (en) * | 2018-08-01 | 2020-02-14 | 珠海格力电器股份有限公司 | Air conditioner control method, device and system |
CN110793148A (en) * | 2018-08-03 | 2020-02-14 | 珠海格力电器股份有限公司 | Air conditioner control method and device |
CN110936969A (en) * | 2018-09-25 | 2020-03-31 | 比亚迪股份有限公司 | Vehicle-mounted air conditioning equipment control method, device and system |
CN111156634A (en) * | 2020-02-24 | 2020-05-15 | 北方工业大学 | Interactive photo-thermal environment control system and method |
CN111189201A (en) * | 2020-01-15 | 2020-05-22 | 西安建筑科技大学 | Air conditioner prediction control method based on machine vision |
CN111351834A (en) * | 2020-03-18 | 2020-06-30 | 南京理工大学 | Intelligent detection device based on magnetic leakage |
CN111351845A (en) * | 2020-03-18 | 2020-06-30 | 南京理工大学 | Intelligent detection device based on pulse eddy current |
CN112506059A (en) * | 2020-12-07 | 2021-03-16 | 常州常工电子科技股份有限公司 | Classroom self-adaptive control system based on energy-saving model of personal sensory comfort |
CN112655007A (en) * | 2018-09-06 | 2021-04-13 | 皇家飞利浦有限公司 | Monitoring movable entities in a predetermined area |
WO2021176700A1 (en) * | 2020-03-06 | 2021-09-10 | 三菱電機株式会社 | Learning device, inference device, and air conditioner |
CN113531784A (en) * | 2020-04-16 | 2021-10-22 | 罗伯特·博世有限公司 | Hvac control refinement granular occupancy pattern estimation |
CN113569739A (en) * | 2021-07-28 | 2021-10-29 | 西安建筑科技大学 | Indoor personnel information identification device and method for auxiliary air supply-on-demand air conditioning system |
CN114459116A (en) * | 2022-01-04 | 2022-05-10 | 珠海格力电器股份有限公司 | Air conditioner control method, air conditioner and nonvolatile storage medium |
CN117928051A (en) * | 2024-03-22 | 2024-04-26 | 山东电工电气集团数字科技有限公司 | Energy-saving carbon reduction control system of central air conditioner |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH08210689A (en) * | 1995-02-07 | 1996-08-20 | Hitachi Plant Eng & Constr Co Ltd | System for estimating air-conditioning thermal load |
JP2005172288A (en) * | 2003-12-09 | 2005-06-30 | Mitsubishi Heavy Ind Ltd | Controlling system for air conditioner |
CN102997374A (en) * | 2012-12-31 | 2013-03-27 | 深圳市奥宇控制系统有限公司 | Method and device for forecasting air-conditioning load and air-conditioner |
CN104008427A (en) * | 2014-05-16 | 2014-08-27 | 华南理工大学 | Central air conditioner cooling load prediction method based on BP neural network |
CN104571034A (en) * | 2014-12-29 | 2015-04-29 | 中国机械工业企业管理协会 | Intelligent management and control equipment for cold and hot joint supply of heating ventilating and air conditioning system and control method thereof |
CN104778503A (en) * | 2015-03-16 | 2015-07-15 | 国家电网公司 | Method for forecasting public building air conditioner short-time base wire load with consideration of real-time weather factors |
CN104848467A (en) * | 2014-02-13 | 2015-08-19 | 海尔集团公司 | Control method of air conditioner and air conditioner |
-
2016
- 2016-04-18 CN CN201610244200.0A patent/CN105910225A/en active Pending
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH08210689A (en) * | 1995-02-07 | 1996-08-20 | Hitachi Plant Eng & Constr Co Ltd | System for estimating air-conditioning thermal load |
JP2005172288A (en) * | 2003-12-09 | 2005-06-30 | Mitsubishi Heavy Ind Ltd | Controlling system for air conditioner |
CN102997374A (en) * | 2012-12-31 | 2013-03-27 | 深圳市奥宇控制系统有限公司 | Method and device for forecasting air-conditioning load and air-conditioner |
CN104848467A (en) * | 2014-02-13 | 2015-08-19 | 海尔集团公司 | Control method of air conditioner and air conditioner |
CN104008427A (en) * | 2014-05-16 | 2014-08-27 | 华南理工大学 | Central air conditioner cooling load prediction method based on BP neural network |
CN104571034A (en) * | 2014-12-29 | 2015-04-29 | 中国机械工业企业管理协会 | Intelligent management and control equipment for cold and hot joint supply of heating ventilating and air conditioning system and control method thereof |
CN104778503A (en) * | 2015-03-16 | 2015-07-15 | 国家电网公司 | Method for forecasting public building air conditioner short-time base wire load with consideration of real-time weather factors |
Non-Patent Citations (3)
Title |
---|
廖传善: "《房间湿度的测量与调节》", 30 April 1987, 中国建筑工业出版社 * |
张嗣瀛: "《2000中国控制与决策学术年会论文集》", 30 April 2000, 东北大学出版社 * |
魏东: "《非线性系统神经网络参数预测及控制》", 31 January 2008, 机械工业出版社 * |
Cited By (31)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US10871302B2 (en) | 2016-12-19 | 2020-12-22 | Lg Electronics Inc. | Artificial intelligence air conditioner and control method thereof |
EP3379163A3 (en) * | 2016-12-19 | 2018-10-31 | Lg Electronics Inc. | Air conditioner and control method thereof |
CN106918119B (en) * | 2017-03-23 | 2019-04-19 | 天津锋尚智慧能源科技发展有限公司 | A kind of central air conditioner system that operation risk is low |
CN106918119A (en) * | 2017-03-23 | 2017-07-04 | 上海巽晔计算机科技有限公司 | A kind of low central air conditioner system of operation risk |
CN106895564A (en) * | 2017-03-27 | 2017-06-27 | 中国科学院广州能源研究所 | A kind of station air conditioner control system and method |
CN107461875A (en) * | 2017-07-07 | 2017-12-12 | 珠海格力电器股份有限公司 | Control method and device of air conditioning system |
CN107883541A (en) * | 2017-10-24 | 2018-04-06 | 珠海格力电器股份有限公司 | Air conditioner control method and device |
CN108413588A (en) * | 2018-02-12 | 2018-08-17 | 北京工业大学 | A kind of personalized air-conditioner control system and method based on thermal imaging and BP neural network |
CN108413588B (en) * | 2018-02-12 | 2021-03-02 | 北京工业大学 | Personalized air conditioner control system and method based on thermal imaging and BP neural network |
CN108664452A (en) * | 2018-05-07 | 2018-10-16 | 北京工业大学 | A kind of indoor occupant number determination method and determining system |
CN108664452B (en) * | 2018-05-07 | 2022-04-26 | 北京工业大学 | Method and system for determining number of indoor personnel |
CN108647817A (en) * | 2018-05-07 | 2018-10-12 | 北京工业大学 | A kind of prediction technique and forecasting system of energy consumption load |
CN108647817B (en) * | 2018-05-07 | 2022-03-25 | 北京工业大学 | Energy consumption load prediction method and system |
CN110793162A (en) * | 2018-08-01 | 2020-02-14 | 珠海格力电器股份有限公司 | Air conditioner control method, device and system |
CN110793148A (en) * | 2018-08-03 | 2020-02-14 | 珠海格力电器股份有限公司 | Air conditioner control method and device |
CN109140665A (en) * | 2018-08-31 | 2019-01-04 | 珠海格力电器股份有限公司 | Method and device for controlling air conditioner and air conditioner device |
CN112655007A (en) * | 2018-09-06 | 2021-04-13 | 皇家飞利浦有限公司 | Monitoring movable entities in a predetermined area |
CN110936969A (en) * | 2018-09-25 | 2020-03-31 | 比亚迪股份有限公司 | Vehicle-mounted air conditioning equipment control method, device and system |
CN109164707A (en) * | 2018-09-28 | 2019-01-08 | 苏州市建筑科学研究院集团股份有限公司 | A kind of indoor environment negative-feedback regu- lation system based on artificial neural network algorithm |
CN111189201A (en) * | 2020-01-15 | 2020-05-22 | 西安建筑科技大学 | Air conditioner prediction control method based on machine vision |
CN111156634A (en) * | 2020-02-24 | 2020-05-15 | 北方工业大学 | Interactive photo-thermal environment control system and method |
WO2021176700A1 (en) * | 2020-03-06 | 2021-09-10 | 三菱電機株式会社 | Learning device, inference device, and air conditioner |
CN111351834A (en) * | 2020-03-18 | 2020-06-30 | 南京理工大学 | Intelligent detection device based on magnetic leakage |
CN111351845A (en) * | 2020-03-18 | 2020-06-30 | 南京理工大学 | Intelligent detection device based on pulse eddy current |
CN113531784A (en) * | 2020-04-16 | 2021-10-22 | 罗伯特·博世有限公司 | Hvac control refinement granular occupancy pattern estimation |
CN112506059A (en) * | 2020-12-07 | 2021-03-16 | 常州常工电子科技股份有限公司 | Classroom self-adaptive control system based on energy-saving model of personal sensory comfort |
CN113569739A (en) * | 2021-07-28 | 2021-10-29 | 西安建筑科技大学 | Indoor personnel information identification device and method for auxiliary air supply-on-demand air conditioning system |
CN114459116A (en) * | 2022-01-04 | 2022-05-10 | 珠海格力电器股份有限公司 | Air conditioner control method, air conditioner and nonvolatile storage medium |
CN114459116B (en) * | 2022-01-04 | 2023-02-17 | 珠海格力电器股份有限公司 | Air conditioner control method, air conditioner and nonvolatile storage medium |
CN117928051A (en) * | 2024-03-22 | 2024-04-26 | 山东电工电气集团数字科技有限公司 | Energy-saving carbon reduction control system of central air conditioner |
CN117928051B (en) * | 2024-03-22 | 2024-05-17 | 山东电工电气集团数字科技有限公司 | Energy-saving carbon reduction control system of central air conditioner |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN105910225A (en) | Air conditioner load control system and method based on personnel information detection | |
Zhang et al. | A review on occupancy prediction through machine learning for enhancing energy efficiency, air quality and thermal comfort in the built environment | |
Peng et al. | Temperature-preference learning with neural networks for occupant-centric building indoor climate controls | |
CN107942960B (en) | A kind of intelligentized information processing system | |
CN107120782B (en) | A kind of HVAC system control method based on multi-user's hot comfort data | |
Li et al. | Experimental study of an indoor temperature fuzzy control method for thermal comfort and energy saving using wristband device | |
Jazizadeh et al. | User-led decentralized thermal comfort driven HVAC operations for improved efficiency in office buildings | |
CN108710947A (en) | A kind of smart home machine learning system design method based on LSTM | |
Jung et al. | Occupant-centered real-time control of indoor temperature using deep learning algorithms | |
CN110377084B (en) | Building indoor environment regulation and control method based on intelligent control strategy | |
CN107085732B (en) | Cowshed environment ammonia intelligent monitoring system based on wireless sensor network | |
CN109634098A (en) | A kind of fattening house environment conditioning system and method | |
CN103890668A (en) | Building automation system control with motion sensing | |
US20200305714A1 (en) | Iot system and monitoring method for monitoring association between indoor environment and health of elderly people | |
CN106614273B (en) | Pet feeding method and system based on Internet of Things big data analysis | |
CN106528986A (en) | Tracking system and method for human body dynamic heat comfort | |
CN106506848A (en) | A kind of livestock culturing control system based on ZigBee | |
CN109883016A (en) | A kind of air pleasant degree adjusting method and equipment | |
EP4128260A2 (en) | System, method and computer program product which uses biometrics as a feedback for home control monitoring to enhance wellbeing | |
CN112613232A (en) | Indoor human body thermal comfort prediction and evaluation method under winter heating condition | |
CN114200839B (en) | Intelligent office building energy consumption control model for dynamic monitoring of coupling environment behaviors | |
CN108426349A (en) | Air-conditioning personalized health management method based on complex network and image recognition | |
CN112734003A (en) | Human skin temperature detection method based on deep convolutional network | |
Kciuk et al. | Design and modeling of intelligent building office and thermal comfort based on probabilistic neural network | |
CN111367163A (en) | Intelligent controller for farm environment and application thereof |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into 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: 20160831 |