CN107678410A - It is a kind of towards the intelligent control method of greenhouse, system and controller - Google Patents

It is a kind of towards the intelligent control method of greenhouse, system and controller Download PDF

Info

Publication number
CN107678410A
CN107678410A CN201710923214.XA CN201710923214A CN107678410A CN 107678410 A CN107678410 A CN 107678410A CN 201710923214 A CN201710923214 A CN 201710923214A CN 107678410 A CN107678410 A CN 107678410A
Authority
CN
China
Prior art keywords
data
control
greenhouse
environmental variance
environmental
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
Application number
CN201710923214.XA
Other languages
Chinese (zh)
Inventor
陈飞
陈一飞
蒲东
李丹
张向南
彭雄
李行健
王亚威
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
China Agricultural University
Original Assignee
China Agricultural University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by China Agricultural University filed Critical China Agricultural University
Priority to CN201710923214.XA priority Critical patent/CN107678410A/en
Publication of CN107678410A publication Critical patent/CN107678410A/en
Pending legal-status Critical Current

Links

Classifications

    • GPHYSICS
    • G05CONTROLLING; REGULATING
    • G05BCONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B19/00Programme-control systems
    • G05B19/02Programme-control systems electric
    • G05B19/418Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS], computer integrated manufacturing [CIM]
    • G05B19/4185Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS], computer integrated manufacturing [CIM] characterised by the network communication
    • G05B19/4186Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS], computer integrated manufacturing [CIM] characterised by the network communication by protocol, e.g. MAP, TOP
    • AHUMAN NECESSITIES
    • A01AGRICULTURE; FORESTRY; ANIMAL HUSBANDRY; HUNTING; TRAPPING; FISHING
    • A01GHORTICULTURE; CULTIVATION OF VEGETABLES, FLOWERS, RICE, FRUIT, VINES, HOPS OR SEAWEED; FORESTRY; WATERING
    • A01G9/00Cultivation in receptacles, forcing-frames or greenhouses; Edging for beds, lawn or the like
    • A01G9/24Devices or systems for heating, ventilating, regulating temperature, illuminating, or watering, in greenhouses, forcing-frames, or the like
    • GPHYSICS
    • G05CONTROLLING; REGULATING
    • G05BCONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B19/00Programme-control systems
    • G05B19/02Programme-control systems electric
    • G05B19/418Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS], computer integrated manufacturing [CIM]
    • G05B19/41885Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS], computer integrated manufacturing [CIM] characterised by modeling, simulation of the manufacturing system
    • YGENERAL 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
    • Y02TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02ATECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
    • Y02A40/00Adaptation technologies in agriculture, forestry, livestock or agroalimentary production
    • Y02A40/10Adaptation technologies in agriculture, forestry, livestock or agroalimentary production in agriculture
    • Y02A40/25Greenhouse technology, e.g. cooling systems therefor

Abstract

Present invention offer is a kind of to be included towards the intelligent control method of greenhouse, system and controller, methods described:S1, using single-factor control mode, multiple environmental variance data in control targe greenhouse, and each environmental variance real data of corresponding record and control implementation procedure data respectively, until record number reaches setting demand;S2, initial neural network model is obtained, and using each environmental variance real data and the control implementation procedure data, train the initial neural network model, obtain control neural network model;S3, based on multiple-factor control process, accordingly multiple environmental variance real data it is expected environmental data with given, utilize the control neural network model, the control data-oriented of the multiple-factor control process is obtained, and the respective environment variable data in the target greenhouse is correspondingly adjusted based on the control data-oriented.The present invention is simple to operate, can effectively reduce resource consumption and improves intelligent level, so as to effectively improve economic benefit.

Description

It is a kind of towards the intelligent control method of greenhouse, system and controller
Technical field
The present invention relates to industrialized agriculture information-oriented production management technical field, more particularly, to one kind towards greenhouse ring Intelligent control method, system and the controller in border.
Background technology
Greenhouse is the space being isolated from the outside built and formed using materials such as glass, films, can make full use of and work as The natural environmental condition on ground.This industrialized agriculture is the effective work for the every monitor and control facility being had by inside, reaches improvement Crop growth environment or the purpose for createing more preferably room climate.Greenhouse is the important component of industrialized agriculture, is existing Most active part in important carrier during generationization agricultural development, and China's agricultural modernization evolution.
At present, green house control mainly has two ways.It is a kind of controlled with single-factor based on, ignore the coupling between envirment factor Conjunction relation, common regulation and control object are temperature or illumination, by temperature either illumination control in constant scope or controller Executing agency is directly controlled to control some envirment factor.The core of this quasi-controller is usually single-chip microcomputer or PLC, control mode Simply, generally require continuous action or keep 24 hours openings, the consumption to the energy is very big, and the control method is applied to It is higher to environmental requirement, and expensive crop.
Another kind is to carry out environmental Kuznets Curves by establishing intelligence system, and executing agency and PC ends are remotely connected, The data that sensor collects are sent to PC ends in real time, and field personnel is judged, the action of control executing agency.Compare In single-factor controller, the control system effectively reduces energy consumption, while realizes the control to multiple executing agencies, But the system can not accomplish intelligentized control method, and not help reducing human cost.In addition, when the communication in PC ends and greenhouse When being out of order, control system can not carry out auto-control.
To sum up, existing Technique for Controlling Greenhouse Environment, intelligent level is relatively low, and resource consumption is big, and overall economic benefit is not It is high.
The content of the invention
In order to overcome above mentioned problem or solve the above problems at least in part, the present invention provides a kind of towards greenhouse Intelligent control method, system and controller, reaching effective streamline operation, reducing resource consumption and improving intelligent water It is flat, so as to effectively improve the purpose of economic benefit.
In a first aspect, the present invention provides a kind of intelligent control method towards greenhouse, including:S1, using single-factor Control mode, respectively multiple environmental variance data in control targe greenhouse, and each environmental variance real data of corresponding record and control Implementation procedure data processed, until record number reaches setting demand, the single-factor control mode refers to respectively with single environment Variable is control targe, individually controls each environmental variance data;S2, initial neural network model is obtained, and utilize each institute Environmental variance real data and the control implementation procedure data are stated, trains the initial neural network model, obtains control god Through network model;S3, based on multiple-factor control process, accordingly multiple environmental variance real data it is expected environmental data, profit with given With the control neural network model, obtain the control data-oriented of the multiple-factor control process, and based on it is described control to Fixed number is according to the corresponding respective environment variable data for adjusting the target greenhouse.
Wherein, each environmental variance real data and the control implementation procedure data, instruction are utilized described in step S2 The step of practicing the initial neural network model further comprises:Using each environmental variance real data as input, correspond to respectively It is described to control implementation procedure data to train the initial neural network model for output.
Wherein, the multiple environmental variance data further comprise gas concentration lwevel, intensity of illumination, temperature and humidity; The control implementation procedure data further comprise, when each environmental variance real data reaches preset value, corresponding environment becomes The duration of amount control implementation procedure.
Further, before the step of S3, methods described also includes:Obtained by remote control terminal described given It is expected environmental data.
Wherein, the remote control terminal further comprises:Industrial computer, cloud server or user's intelligent terminal.
Second aspect, the present invention provide a kind of intelligence control system towards greenhouse, including:It is greenhouse controller, more The environmental data sensor of multiple environmental variance real data, described in variable environment adjustment actuating mechanism and measurement target greenhouse Greenhouse controller communicates to connect between each environmental data sensor and each environment adjustment actuating mechanism respectively;It is described Greenhouse controller is used for:Using single-factor control mode, the multiple environmental variance data, and each ring of corresponding record are controlled respectively Border variable real data and control implementation procedure data, until record number reaches setting demand, the single-factor control mode Refer to respectively using single environmental variance as control targe, individually control each environmental variance data;Build initial neutral net Model, and using each environmental variance real data and the control implementation procedure data, train the initial neutral net Model, obtain control neural network model;And based on multiple-factor control process respective environment variable real data and to regular Environmental data is hoped, using the control neural network model, obtains the control data-oriented of the multiple-factor control process, and base The respective environment variable number in the target greenhouse is adjusted in environment adjustment actuating mechanism corresponding to the control data-oriented control According to.
Further, the system also includes:Connect the remote control terminal of the greenhouse controller, the remote control Terminal is used to obtain and issues the given expectation environmental data to the greenhouse controller.
Wherein, the greenhouse controller further comprises wireless communication module and/or wire communication module, the long-range control Terminal processed further comprises wireless communication module and/or wire communication module, the greenhouse controller and remote control end Communicated between end using udp protocol;The remote control terminal issues time interval to the greenhouse controller by setting The given expectation environmental data is issued, the given expectation environmental data includes each environmental variance optimal value and each environmental variance Value range;The greenhouse controller uploads the multiple-factor to the remote control terminal by setting uplink time interval and controlled Journey respective environment variable real data.
Wherein, the greenhouse controller is further additionally operable to:When judging with the remote control terminal communicating interrupt, press The given expectation environmental data that last time receives, automatically controls the environmental variance data in the target greenhouse, and by institute State multiple-factor control process respective environment variable real data and be stored in the data memory module, until whole with the remote control The communication at end recovers, and switches to multiple-factor control process.
The third aspect, the present invention provide a kind of intelligent controller towards greenhouse, including:Central processing unit and with Data memory module, communication module, analog-to-digital converting module and the relay mould that the central processing unit communicates to connect respectively Block;The environmental data sensing of multiple environmental variance real data in the analog-to-digital converting module connection measurement target greenhouse The output end of device, the relay module connect the input of each environment adjustment actuating mechanism in the target greenhouse;The number It is used to store towards the intelligent control algorithm of greenhouse, environmental variance real data, control implementation procedure number according to memory module According to, multiple-factor control process respective environment variable real data, given it is expected environmental data and control data-oriented;The center Processor obtains each environmental variance real data by the analog-to-digital converting module, and is used for:Using single-factor control Mode processed, the multiple environmental variance data are controlled respectively, and each environmental variance real data of corresponding record and control performed Number of passes evidence, until record number reaches setting demand, the single-factor control mode refers to respectively using single environmental variance as control Target processed, individually control each environmental variance data;Initial neural network model is built, and it is real using each environmental variance Border data and the control implementation procedure data, train the initial neural network model, obtain control neural network model;With And based on the multiple-factor control process respective environment variable real data and the given expectation environmental data, using described Control neural network model, the control data-oriented of the multiple-factor control process is obtained, and will by the relay module The control data-oriented is issued to corresponding environment adjustment actuating mechanism, adjusts the respective environment variable number in the target greenhouse According to.
It is provided by the invention a kind of towards the intelligent control method of greenhouse, system and controller, for greenhouse The nonlinear characteristic of multiple-factor coupling, multiple-input and multiple-output Controlling model is built using based on neural network algorithm, by temperature Indoor ambient parameter (temperature, humidity, CO2Concentration and intensity of illumination) carry out real time automatic detection, Intelligent adjustment greenhouse. In addition, by setting remote control terminal, controlled available for greenhouse group system, the management and control to ten thousand greenhouses is realized, when single During communicating interrupt between greenhouse controller and remote control terminal, greenhouse controller can carry out autonomous control to greenhouse.This Invent it is simple to operate, stable and reliable in work, practical, can effectively reduce resource consumption and improve intelligent level, so as to Effectively improve economic benefit.
Brief description of the drawings
Fig. 1 is a kind of flow chart of intelligent control method towards greenhouse of the embodiment of the present invention;
Fig. 2 is a kind of training managing process flow diagram flow chart of neutral net of the embodiment of the present invention;
Fig. 3 is a kind of overall architecture schematic diagram of intelligence control system towards greenhouse of the embodiment of the present invention;
Fig. 4 is a kind of control system overall architecture schematic diagram of the embodiment of the present invention;
Fig. 5 is a kind of greenhouse clustered control structure composed schematic diagram of the embodiment of the present invention;
Fig. 6 is a kind of network interface communicating circuit structural representation of the embodiment of the present invention;
Fig. 7 is a kind of structured flowchart of intelligent controller towards greenhouse of the embodiment of the present invention;
Fig. 8 is a kind of overall structure diagram of intelligent controller of the embodiment of the present invention;
Fig. 9 is a kind of AI interface circuit figures of intelligent controller of the embodiment of the present invention;
Figure 10 is a kind of plug type AI interface terminals subgraphs of intelligent controller of the embodiment of the present invention;
Figure 11 is a kind of pluggable DO interface circuit figures of intelligent controller of the embodiment of the present invention;
Figure 12 is that a kind of hand push button of intelligent controller of the embodiment of the present invention connects circuit diagram;
Figure 13 is that a kind of buzzer of intelligent controller of the embodiment of the present invention connects circuit diagram;
Figure 14 is a kind of LCD interface circuit figures of intelligent controller of the embodiment of the present invention;
Figure 15 is a kind of runnable interface screenshotss schematic diagram of Intelligent Greenhouse Controller of the embodiment of the present invention.
Embodiment
To make the object, technical solutions and advantages of the present invention clearer, below in conjunction with attached in the embodiment of the present invention Figure, the technical scheme in the present invention is clearly and completely described, it is clear that described embodiment is one of the present invention Divide embodiment, rather than whole embodiments.Based on the embodiment in the present invention, those of ordinary skill in the art are not making The every other embodiment obtained on the premise of creative work, belongs to the scope of protection of the invention.
As the one side of the embodiment of the present invention, the present embodiment provides a kind of intelligent control side towards greenhouse Method, it is a kind of flow chart of intelligent control method towards greenhouse of the embodiment of the present invention with reference to figure 1, including:
S1, using single-factor control mode, respectively multiple environmental variance data in control targe greenhouse, and corresponding record is each Environmental variance real data and control implementation procedure data, until record number reaches setting demand, the single-factor controlling party Formula refers to respectively using single environmental variance as control targe, individually controls each environmental variance data.
It is to be understood that running the starting stage in controller, the coupling of multiple environmental variances in target greenhouse is not considered first Relation, respectively using each single environmental variance in target greenhouse as control targe, individually control each environmental variance data.In real time Gather and record each environmental variance real data in control process, while corresponding record is issued to the control of control requirement in the data The control implementation procedure data of process processed.The confining spectrum of controller operation starting stage is to be put into operation since controller, Reach the period of setting demand to data acquisition number.
It is wherein optional, the multiple environmental variance data further comprise gas concentration lwevel, intensity of illumination, temperature and Humidity;The control implementation procedure data further comprise, when each environmental variance real data reaches preset value, corresponding ring The duration of border Variable Control implementation procedure.
It is to be understood that greenhouse is collectively constituted by multinomial environment parameter, coupling between each environment parameter be present Relation, each environment parameter data are adjusted by associating, reach the greenhouse condition of needs.The present embodiment considers greenhouse Gas concentration lwevel, intensity of illumination, temperature and humidity parameter, the numerical value that four environmental variances are adjusted by associating reach needs Greenhouse condition.
Assuming that setting times of collection as N, the starting stage is run in controller, greenhouse controller is in a manner of single-factor controls Work, when some amount or several amounts exceed preset range, during the action of each executing agency of greenhouse controller setting Between ti=0, goal of regulation and control is the optimal value of corresponding amount, at the end of regulation and control, records regulation and control time, the time corresponding to each amount As control implementation procedure data.
S2, initial neural network model is obtained, and performed using each environmental variance real data and the control Number of passes evidence, the initial neural network model is trained, obtain control neural network model.
It is to be understood that this step realizes foundation and the training process of target control neural network model.It is determined that input, Output variable information, and according to input, after output variable information establishes initial neural network model, initial neural network model Certain primary data is needed to be trained, but greenhouse is relative complex, collection and the screening for carrying out data are relatively more tired It is difficult.The present embodiment is trained the collection and training of data in the starting stage that controller is run.
It is wherein optional, each environmental variance real data and the control implementation procedure number are utilized described in step S2 According to, train the initial neural network model the step of further comprise:Using each environmental variance real data as input, it is right Answer and each described control implementation procedure data to train the initial neural network model for output.
The input and output of neutral net can represent that input vector is designated as X=(x with two one-dimensional vectors1,x1,K, xn), output vector is designated as T=(t1,t1,K,tm).Wherein x1、x2、...、xnN input of respectively neural network model becomes Amount, t1、t2、...、tnRespectively m output variable of neural network model.
For example, for the gas concentration lwevel in above-mentioned steps, intensity of illumination, temperature and humidity parameter, if neutral net Input variable include the actual gas concentration lwevel of greenhouse, intensity of illumination, temperature and humidity parameter, and setting dioxy Change concentration of carbon, intensity of illumination and temperature value;Output variable includes the execution time of regulation of carbon dioxide concentration executing agency, regulation Execution time, the execution time of regulation temperature actuator and the holding for regulation humidity executing agency of intensity of illumination executing agency The row time.The input vector for then building neural network model is X=(x1,x1,K,x7), output vector is T=(t1,t1,K,t4)。
Regulation and control terminate, it is necessary to corresponding input and output are normalized, the present embodiment uses sigmoid every time Function is normalized, and formula is as follows:
After training sample collection capacity reaches setting value, the flow according to Fig. 2 is carried out to initial neural network model Training, Fig. 2 are a kind of training managing process flow diagram flow chart of neutral net of the embodiment of the present invention, including:To current group training sample It is normalized, and the neural network algorithm meter of initial neural network model is carried out to the training sample after normalized Calculate;Judge that all training samples whether using finishing, if so, then terminating training program, are otherwise chosen next group of training sample and entered Row neural network algorithm calculates;Carry out forward calculation and whether error in judgement rate meets given threshold, if so, then judging all instructions Whether practice sample using finishing, otherwise carry out backpropagation, update neural network weight and threshold value, continue forward calculation, Until error rate reaches given threshold.
S3, based on multiple-factor control process, accordingly multiple environmental variance real data it is expected environmental data with given, utilize The control neural network model, the control data-oriented of the multiple-factor control process is obtained, and it is given based on the control Data correspondingly adjust the respective environment variable data in the target greenhouse.
It is to be understood that according to collection of the above-mentioned steps completion to training sample data and to initial neural network model Training after, into multiple-factor control process.In multiple-factor control process, multiple-factor control is first gathered by environmental data sensor The environmental variance real data of process processed, and the given expectation environmental data that rule of thumb algorithm is calculated is obtained, this is given It is expected that environmental data can include greenhouse optimal value or term of reference etc..
By the environmental variance real data of multiple-factor control process and given expectation environmental data together as control nerve The input variable of network, calculated through control neural network Model coupling, obtain the control data-oriented of control greenhouse data, The data-oriented corresponds to multiple governor motions.Then environmental variance governor motion corresponding to control exports according to neural network model Control data-oriented carry out greenhouse variable regulation.
A kind of intelligent control method towards greenhouse provided in an embodiment of the present invention, for greenhouse multiple-factor coupling The nonlinear characteristic of conjunction, multiple-input and multiple-output Controlling model is built using based on neural network algorithm, by the ring in greenhouse Border parameter carries out real time automatic detection, Intelligent adjustment greenhouse.Because neutral net can preferably simulate complex environment, no With coupled relation complicated between the computing environment factor again;Executing agency need not be always maintained at opening, simple to operate, work Make it is reliable and stable, can effectively reduce resource consumption and improve intelligent level, so as to effectively improve economic benefit.
Further, before the step of S3, methods described also includes:Obtained by remote control terminal described given It is expected environmental data.
It is to be understood that in multiple-factor control process, greenhouse controller is needed according to actual environment data and to regular Hope environmental data carry out neural network algorithm calculating, so as to adjusted greenhouse variable data controlled quentity controlled variable, that is, control to Fixed number evidence.Acted by environmental variance governor motion corresponding to control by control data-oriented, the reality in adjustment target greenhouse Environmental data, actual environment data value is set to tend to given environmental data.
Given expectation environmental data therein is typically to be calculated to obtain according to factors such as agrotypes in greenhouse, or according to warp Test the data of plant growth or optimal span in the most suitable target greenhouse of acquisition.Usual warm indoor crops And plant growth conditions are not changeless, therefore the given expectation environmental data in greenhouse controller is to need real-time update 's.
Calculated and obtained by remote control terminal, or by inputting remote control terminal by manually rule of thumb setting, then Sent by remote control terminal to greenhouse controller, target greenhouse variable data is correspondingly adjusted for greenhouse controller.
Meanwhile for each remote control terminal, control instruction can be sent to more greenhouse controllers, realized to temperature The centralized management of room cluster.
In one embodiment, the remote control terminal further comprises:Industrial computer, cloud server or user's intelligence Terminal.
It is to be understood that given it is expected that environmental data can either cloud server or user hold by neighbouring industrial computer Some intelligent mobile terminals obtain and are sent to target greenhouse controller.
A kind of intelligent control method towards greenhouse provided in an embodiment of the present invention, by by remote control terminal not Suitable environment parameter is given in disconnected renewal greenhouse, and greenhouse data can be made to be always held at the optimal model of suitable plant growth Enclose, meet that chamber crop grows demand, produce higher economic benefit, and for greenhouse cluster, it is possible to achieve to more greenhouses Unified intelligent control.
As the other side of the embodiment of the present invention, the present embodiment provides a kind of intelligent control system towards greenhouse System, it is a kind of overall architecture schematic diagram of intelligence control system towards greenhouse of the embodiment of the present invention with reference to figure 3, including: The environment of multiple environmental variance real data in greenhouse controller 1, multivariable environment adjustment actuating mechanism 2 and measurement target greenhouse Data pick-up 3.
Wherein, greenhouse controller 1 communicates between each environmental data sensor 3 and each environment adjustment actuating mechanism 2 respectively Connection.Greenhouse controller 1 is used for:Using single-factor control mode, the multiple environmental variance data, and corresponding note are controlled respectively Each environmental variance real data and control implementation procedure data are recorded, until record number reaches setting demand, the single-factor control Mode processed refers to respectively using single environmental variance as control targe, individually controls each environmental variance data;The initial god of structure Through network model, and using each environmental variance real data and the control implementation procedure data, train the initial god Through network model, control neural network model is obtained;And based on multiple-factor control process respective environment variable real data and It is given it is expected environmental data, using the control neural network model, the control of the multiple-factor control process is obtained to fixed number According to, and based on the respective rings for controlling environment adjustment actuating mechanism 2 corresponding to data-oriented control to adjust the target greenhouse Border variable data.
It is to be understood that from the point of view of by system architecture, control system includes three parts, i.e. information gathering part, control Device part and executing agency part processed, while controller part and information gathering part and executing agency part pass through communicate respectively Realize communication connection in part.Control system integrated stand composition refers to.
Information gathering part is mainly made up of environmental data sensor 3, including carbon dioxide (CO2) sensor, temperature pass Sensor, humidity sensor and optical sensor etc..The monitoring of greenhouse is completed by environmental data sensor 3, environmental data Sensor 3 monitors greenhouse environment parameter in real time, and in the certain time of setting, environmental data sensor 3 carries out a data and sent. The data of transmission are the data gathered recently, because there is multiple sensor nodes in greenhouse, the data of each node collection It is different from, therefore, inside controller needs to handle data, calculates the numerical value of a determination.
Greenhouse area is larger, and monitoring to whole greenhouse can not be completed by relying solely on one group of environmental data sensor 3, Therefore, it is necessary to which structure according to greenhouse, is divided into several pieces of regions, each piece of region sets identical sensor network by greenhouse Network is monitored.
Executing agency part is made up of environment adjustment actuating mechanism 2, and executing agency is to the output according to greenhouse controller 1 Data-oriented is controlled to perform adjustment action.Environment adjustment actuating mechanism 2 connects greenhouse by passing sequentially through contactor with relay Controller 1.Environment adjustment actuating mechanism 2 is driven by contactor, and driven object includes heater, blower fan, CO2Release, Alarm, vent window, insulation quilt and illumination facilities etc., the DO mouths of greenhouse controller 1 are by the break-make of control relay and continue Time controls the motor action of environment adjustment actuating mechanism 2.
Controller part is made up of greenhouse controller 1, is run the starting stage in greenhouse controller 1, is not first considered target temperature The coupled relation of multiple environmental variances in room, greenhouse controller 1 is respectively using each single environmental variance in target greenhouse as control mesh Mark, individually controls each environmental variance data.
Meanwhile greenhouse controller 1 gathers in real time and records each environmental variance real data in control process, and corresponding record The control implementation procedure data for the control process that control requires are issued in the data.Greenhouse controller 1 runs the starting stage Confining spectrum is to be put into operation since greenhouse controller 1, and the period of setting demand is reached to data acquisition number.
Greenhouse controller 1 also realizes foundation and the training process of target control neural network model.It is determined that input, output Variable information, and according to input, after output variable information establishes initial neural network model, greenhouse controller 1 is needed to first Beginning neural network model is trained with certain primary data, but greenhouse is relative complex, carries out the collection of data It is relatively difficult with screening.The present embodiment is trained the collection and training of data in the starting stage that greenhouse controller 1 is run.When After training sample collection capacity reaches setting value, greenhouse controller 1 is trained to initial neural network model, obtains the control of needs Neural network model processed.
After neural network model trains, greenhouse controller 1 is transferred to the intelligent control stage.When the environment in greenhouse becomes Change, during beyond default scope, neural network model is started working.Opening timing device, and timer is set to 0, now, foundation is prolonged When function, the data are fitted into function, the execution time of the function, are exactly that result of calculation conversion respective execution mechanisms continue Time.
Before timer execution, DO mouths put 1, and relay conducting, executing agency acts.Timer timing terminates, and DO mouths are put 0, now relay be no longer turned on.After execution, wait for a period of time, then the data that sensor collects judged, Judge whether greenhouse meets preset requirement, if do not met, continue to adjust.Specific conducting flow with reference to figure 4, For a kind of control system overall architecture schematic diagram of the embodiment of the present invention.
A kind of intelligence control system towards greenhouse provided in an embodiment of the present invention, for greenhouse multiple-factor coupling The nonlinear characteristic of conjunction, by setting greenhouse controller 1 to build multiple-input and multiple-output control mould using based on neural network algorithm Type, real time automatic detection, Intelligent adjustment greenhouse are carried out to the ambient parameter in greenhouse.Because neutral net can be preferable Complex environment is simulated, without coupled relation complicated between the computing environment factor again;Executing agency need not be always maintained at opening State, it is simple to operate, stable and reliable in work, can effectively reduce resource consumption and improve intelligent level, so as to effectively improve Economic benefit.
Further, the system also includes:Connect the remote control terminal of the greenhouse controller, the remote control Terminal is used to obtain and issues the given expectation environmental data to the greenhouse controller.
It is to be understood that the greenhouse controller 1 of the present embodiment is the controller towards remote control terminal, it is greenhouse cluster The control towards every greenhouse layer of control system.Remote control terminal can use local industrial computer, long-range cloud server Or user's hand-held intelligent terminal.Remote control terminal calculates according to factors such as agrotypes in greenhouse obtains given expectation environment The data of plant growth or optimal span in data, or the most suitable target greenhouse rule of thumb obtained.It Remote control terminal sends the regular prestige environmental data of acquisition to greenhouse controller 1 afterwards, for the corresponding regulation mesh of greenhouse controller 1 Mark greenhouse variable data.
In addition, a remote control terminal can control multiple greenhouse controllers 1, responsible pair of each greenhouse controller 1 simultaneously The specific control in greenhouse is answered, the present embodiment remote control terminal realizes that the structure composed of greenhouse clustered control is this hair with reference to figure 5 A kind of bright greenhouse clustered control structure composed schematic diagram of embodiment.Remote control terminal issues control instruction, greenhouse controller 1 The ambient parameter in greenhouse is controlled according to control instruction, and the environmental data monitored in real time is sent to remote control terminal.
A kind of intelligence control system towards greenhouse provided in an embodiment of the present invention, by setting remote control whole End, realize that constantly updating suitable environment parameter in greenhouse gives, and can make greenhouse data be always held at suitable crop life Long optimum range, meet that chamber crop grows demand, produce higher economic benefit, and for greenhouse cluster, it is possible to achieve To the unified intelligent control in more greenhouses.
Wherein optional, the greenhouse controller further comprises wireless communication module and/or wire communication module, described Remote control terminal further comprises wireless communication module and/or wire communication module, the greenhouse controller with it is described long-range Communicated between control terminal using udp protocol;The remote control terminal issues time interval to the greenhouse by setting Controller issues the given expectation environmental data, and the given expectation environmental data includes each environmental variance optimal value and each ring Border range of variables value;The greenhouse controller uploads the multiple-factor by setting uplink time interval to the remote control terminal Control process respective environment variable real data.
It is to be understood that because in greenhouse clustered control, a remote control terminal is logical with multiple greenhouse controllers 1 Letter, in order to ensure communication efficiency and traffic rate, using udp protocol, greenhouse controller 1 and the communication modes of green house control terminal There are two kinds of selections, wireless telecommunications or wire communication.Greenhouse controller 1 is provided with network interface, is entered by optical fiber and remote control terminal Row communication.
Meanwhile the inside of greenhouse controller 1 is provided with WIFI module and 4G modules, remote control terminal is logical with greenhouse controller 1 4G/WIFI communications are crossed, i.e., greenhouse controller 1 can also be communicated wirelessly with remote control terminal, Neng Gouyou The scope of application of effect extension greenhouse controller.
It is a kind of network interface communicating circuit structural representation of the embodiment of the present invention with reference to figure 6 by taking network interface communicating circuit as an example, The core component of network interface communicating circuit is LAN8720A, and the information received is converted to rs 232 serial interface signal by network port circuit, with string Degree of lip-rounding formula is sent in processor.
Week about, remote control terminal issues a control instruction to greenhouse controller 1, and the instruction is included in two aspects Hold:The optimal value x_ of greenhouse parameterbestWith term of reference value (xmin,xmax), the target of greenhouse controller 1 is by greenhouse ring Border is maintained in default scope.Uploaded in addition, greenhouse controller 1 every two hours carries out a data, upload the content of data For the current environmental information in greenhouse.
A kind of intelligence control system towards greenhouse provided in an embodiment of the present invention, by remote control terminal with The design of communication protocol and communication mode between greenhouse controller, can ensure communication efficiency and traffic rate.
Wherein optional, the greenhouse controller is further additionally operable to:In judging to communicate with the remote control terminal When disconnected, the given expectation environmental data that is received by last time, the environmental variance data in the target greenhouse are automatically controlled, And the multiple-factor control process respective environment variable real data is stored in the data memory module, until with it is described long-range Control the communication of terminal to recover, switch to multiple-factor control process.
It is to be understood that greenhouse controller 1 and remote control terminal be by network communication, but greenhouse controller 1 and long-range Communication between control terminal may be because that some reasons are interrupted.Under normal circumstances, greenhouse controller 1 is to remote control terminal Data are sent, after remote control terminal receives data, signal is finished to the feedback reception of greenhouse controller 1.In greenhouse controller 1 After having sent data, wait receives signal and starts timing.
After beyond the default time, greenhouse controller 1 sends data again, if not receiving feedback, green house control also Device 1 is determined as the communicating interrupt with remote control terminal.Now, greenhouse controller 1 carries out automatic control mode, and environment is maintained In the range of newest setting, the data that sensor collects do not retransmit, and are stored temporarily in local database.
After the communication of greenhouse controller 1 and remote control terminal is normal, when remote control terminal issues extraction signal, temperature Chamber controller 1 sends data, and wait receives signal, after receiving, empties the data of storage.Now, green house control Device 1 is considered as communication normally, receives the control instruction of remote control terminal, and greenhouse is controlled according to control instruction.
A kind of intelligence control system towards greenhouse provided in an embodiment of the present invention, by setting greenhouse controller It from control model, can ensure after the communication disruption with remote control terminal, continue to carry out greenhouse according to newest data-oriented Environmental Kuznets Curves, system crash can be effectively avoided, brings economic loss.
As the another aspect of the embodiment of the present invention, the present embodiment provides a kind of intelligent control towards greenhouse Device, it is a kind of structured flowchart of intelligent controller towards greenhouse of the embodiment of the present invention with reference to figure 7, including:Central processing Device 101 and the data memory module 102 communicated to connect respectively with central processing unit 101, communication module 103, analog/digital turn Change the mold block 104 and relay module 105.
Wherein, the connection of analog-to-digital converting module 104 measures the environment of multiple environmental variance real data in target greenhouse The output end of data pick-up, relay module 105 connect the input of each environment adjustment actuating mechanism in the target greenhouse; Data memory module 102 is used to store to be performed towards the intelligent control algorithm of greenhouse, environmental variance real data, control Number of passes evidence, multiple-factor control process environmental variance real data, given expectation environmental data and control data-oriented.
Central processing unit 101 obtains each environmental variance real data by analog-to-digital converting module 104, is used in combination In:
Using single-factor control mode, the multiple environmental variance data, and each environmental variance of corresponding record are controlled respectively Real data and control implementation procedure data, until record number reaches setting demand, the single-factor control mode refers to point Not using single environmental variance as control targe, each environmental variance data are individually controlled;
Initial neural network model is built, and utilizes each environmental variance real data and the control implementation procedure number According to, the training initial neural network model, acquisition control neural network model;
And based on the multiple-factor control process respective environment variable real data and the given expectation environment number According to, using the control neural network model, obtain the control data-oriented of the multiple-factor control process, and by it is described after The control data-oriented is issued to corresponding environment adjustment actuating mechanism by electrical appliance module, adjusts the corresponding of the target greenhouse Environmental variance data.
It is to be understood that controller during green house control is performed, it is necessary to according to greenhouse actual environment data and give Determine expected data calculating control to give, the given corresponding execution unit of needs of control of acquisition, which corresponds to, to be performed, while needs are to control The data of process processed are stored.Therefore need to set corresponding functional unit in the controller.Specifically include:For obtaining ring It is the analog-to-digital converting module 104 of border data, the central processing unit 101 calculated for control algolithm, fixed for passing control to The relay module 105 of data, the data memory module 102 for storage control process data and for being communicated with relevant device Communication module 103.
In one embodiment, central processing unit uses arm processor, sets 8 simulation inputs (AI) on the controller Interface and 8 numeral output (DO) interfaces, with reference to figure 8, the overall structure for a kind of intelligent controller of the embodiment of the present invention is shown It is intended to, AI interfaces are used for external sensor, and 2 RS485 serial ports and WIFI module are used for external other sensors or sensing Device node, the data of AI interfaces collection are changed by 16 built-in A/D modular converters, and data are sent to central processing unit GPIO port, the port is arranged to input.
The reading for carrying out sensing data in a program is set, and every 30s, the GPIO mouths port of 8 connection AI mouths is entered Row single pass, read data.RS485 communication modes are serial communication, when in use, can by external RS485 expanders With external multiple equipment, single RS485 serial ports could support up external 32 nodes.WIFI module can be used for and sensor section Point or sensor are communicated, and WIFI module can be with real-time reception sensing data, but data is not preserved, in greenhouse control Device processed just preserved during digital independent.
The Interface design circuit of AI interfaces is as shown in figure 9, Fig. 9 is a kind of AI interfaces of intelligent controller of the embodiment of the present invention Circuit diagram.The A/D conversion elements that the present embodiment uses are AD7699, and the element has 8 input ports, and the value of input is adopted for sensor The magnitude of voltage collected, magnitude of voltage and reference voltage compare, and real output valve is obtained according to this.
For the ease of repair and maintenance, AI interfaces use plug-in terminal, the design circuit knot of terminal in one embodiment Structure is as shown in Figure 10, and Figure 10 is a kind of plug type AI interface terminals subgraphs of intelligent controller of the embodiment of the present invention.
The present embodiment sets 8 DO interfaces on the controller, and corresponding environment adjustment actuating mechanism is controlled with corresponding.DO ends Mouthful voltage is 5V, is not directly disposed as external terminal, each external relay in DO ports, when DO puts 1 in port, relay Device turns on, now the external terminal conducting of relay.
Environment adjustment actuating mechanism power is larger, and the external DC24 of terminal of relay turns AC220 relays, should by control Relay turns on, and then the conducting of control contactor, final to drive executing agency's action.In addition to easy access is safeguarded, In one embodiment, DO interfaces use plug type structure, circuit structure such as Figure 11 institutes of plug type structure DO terminal interfaces Show, Figure 11 is a kind of plug type DO interface circuit figures of intelligent controller of the embodiment of the present invention.
In another embodiment, in order to prevent some contingencies from causing greenhouse controller cisco unity malfunction, controlling Manual operation function, the manually operated executing agency that can directly control in greenhouse are added on device processed.The manual behaviour of controller Make, by button completion, button have been added on controller housing, manual operation button is directly connected on relay, when pressing During operation button, relay directly turns on, and then controls executing agency's action.The circuit structure of button is as shown in figure 12, Figure 12 Circuit diagram is connected for a kind of hand push button of intelligent controller of the embodiment of the present invention.
In yet another embodiment, in order to prevent executing agency's perseveration in greenhouse, in the controller, to each ring Border controling parameter is provided with corresponding adjustable range.If some ambient parameter in greenhouse is not up to optimal value, but in phase In the range of answering, controller will not send action command.
In yet another embodiment, it is contemplated that controller is in auto-control, in fact it could happen that because the damage of executing agency Or faulty sensor causes controller can not complete the situation of regulation and control, sets time of fire alarming on the controller.The control of controller System is completed by DO mouths, after the completion of the control of a DO mouth, monitoring period t is set, in monitoring period t, if the equipment Action, i.e., there is level change in DO mouths, then start to be monitored it.If DO mouths level change continuously occurs reaching setting time Number, such as three times, then controller stopping act, and send warning message to remote control terminal.
In addition, also setting alarm lamp on the controller, the triggering of alarm lamp is under software control, is breaking down or mistake Afterwards, manage alarm lamp and the GPIO mouths of buzzer put 1, control the triode ON of the two, alarm lamp lights, and buzzer persistently moves Make.Only controller is switched to after manual mode and can just stopped by staff, and the control circuit of buzzer is as shown in figure 13, figure 13 connect circuit diagram for a kind of buzzer of intelligent controller of the embodiment of the present invention.The control circuit of alarm lamp controls with buzzer Circuit is identical.
In yet another embodiment, set controller that there is screen display function.By setting LCD to show in the controller Screen, realize and operation interface and information displaying interface are set.The hardware circuit of display screen is as shown in figure 14, and Figure 14 is implemented for the present invention A kind of LCD interface circuit figures of intelligent controller of example.
The content of LCD display displaying includes inquiry control record, the situation for inquiring about current greenhouse and current temperature Ecotopia situation of room etc..With reference to figure 15, show for a kind of runnable interface screenshotss of Intelligent Greenhouse Controller of the embodiment of the present invention It is intended to, shows that the controller of the embodiment of the present invention can be to the temperature of greenhouse, humidity, illuminance, dense carbon dioxide in figure The input quantity such as degree, soil conductivity and soil moisture content measures and shows, while to top portion ventilation window motor, bottom ventilation Window motor, volume by motor, carbon dioxide release, plant root warmer, top portion ventilation window close, bottom ventilation window close and Volume is monitored and shown by motor mobile phone actuator.
A kind of intelligent controller towards greenhouse provided in an embodiment of the present invention, coupled for greenhouse multiple-factor Nonlinear characteristic, using based on neural network algorithm build multiple-input and multiple-output Controlling model, by the environment in greenhouse Parameter carries out real time automatic detection, Intelligent adjustment greenhouse.It is simple to operate, stable and reliable in work, can effectively reduce resource Consumption and raising intelligent level, so as to effectively improve economic benefit.
Finally it should be noted that:The above embodiments are merely illustrative of the technical solutions of the present invention, rather than its limitations;Although The present invention is described in detail with reference to the foregoing embodiments, those skilled in the art should be understood:It still can be right Technical scheme described in foregoing embodiments is modified, or carries out equivalent substitution to which part technical characteristic;And this A little modifications are replaced, and the essence of appropriate technical solution is departed from the spirit and model of various embodiments of the present invention technical scheme Enclose.

Claims (10)

  1. A kind of 1. intelligent control method towards greenhouse, it is characterised in that including:
    S1, using single-factor control mode, respectively multiple environmental variance data in control targe greenhouse, and each environment of corresponding record Variable real data and control implementation procedure data, until record number reaches setting demand, the single-factor control mode is Refer to respectively using single environmental variance as control targe, individually control each environmental variance data;
    S2, initial neural network model is obtained, and utilize each environmental variance real data and the control implementation procedure number According to, the training initial neural network model, acquisition control neural network model;
    S3, based on multiple-factor control process accordingly multiple environmental variance real data and it is given it is expected environmental data, using described Control neural network model, the control data-oriented of the multiple-factor control process is obtained, and be based on the control data-oriented The corresponding respective environment variable data for adjusting the target greenhouse.
  2. 2. according to the method for claim 1, it is characterised in that each environmental variance actual number is utilized described in step S2 Further comprise according to described the step of controlling implementation procedure data, training the initial neural network model:
    By input of each environmental variance real data, it is corresponding it is each it is described to control implementation procedure data be output, described in training Initial neural network model.
  3. 3. according to the method for claim 1, it is characterised in that:
    The multiple environmental variance data further comprise gas concentration lwevel, intensity of illumination, temperature and humidity;
    The control implementation procedure data further comprise, when each environmental variance real data reaches preset value, corresponding ring The duration of border Variable Control implementation procedure.
  4. 4. according to any described method in claims 1 to 3, it is characterised in that before the step of S3, in addition to:
    The given expectation environmental data is obtained by remote control terminal.
  5. 5. according to the method for claim 4, it is characterised in that the remote control terminal further comprises:
    Industrial computer, cloud server or user's intelligent terminal.
  6. A kind of 6. intelligence control system towards greenhouse, it is characterised in that including:Greenhouse controller, multivariable environment are adjusted Save executing agency and measure the environmental data sensor of multiple environmental variance real data in target greenhouse, the greenhouse controller Communicated to connect respectively between each environmental data sensor and each environment adjustment actuating mechanism;
    The greenhouse controller is used for:
    Using single-factor control mode, the multiple environmental variance data are controlled respectively, and each environmental variance of corresponding record is actual Data and control implementation procedure data, until record number reaches setting demand, the single-factor control mode refer to respectively with Single environmental variance is control targe, individually controls each environmental variance data;
    Initial neural network model is built, and utilizes each environmental variance real data and the control implementation procedure data, The initial neural network model is trained, obtains control neural network model;
    And based on multiple-factor control process respective environment variable real data and given expectation environmental data, utilize the control Neural network model processed, the control data-oriented of the multiple-factor control process is obtained, and based on the control data-oriented control Environment adjustment actuating mechanism corresponding to system adjusts the respective environment variable data in the target greenhouse.
  7. 7. system according to claim 6, it is characterised in that also include:Connect the remote control of the greenhouse controller Terminal, the remote control terminal are used to obtain and issue the given expectation environmental data to the greenhouse controller.
  8. 8. system according to claim 7, it is characterised in that the greenhouse controller further comprises wireless communication module And/or wire communication module, the remote control terminal further comprise wireless communication module and/or wire communication module, institute State and communicated between greenhouse controller and the remote control terminal using udp protocol;
    The remote control terminal issues time interval by setting and issues the given expectation environment number to the greenhouse controller According to the given expectation environmental data includes each environmental variance optimal value and each environmental variance value range;
    The greenhouse controller uploads the multiple-factor control process by setting uplink time interval to the remote control terminal Respective environment variable real data.
  9. 9. system according to claim 8, it is characterised in that the greenhouse controller is further additionally operable to:
    When judge with the remote control terminal communicating interrupt when, by last time receive the given expectation environmental data, Automatically control the environmental variance data in the target greenhouse, and by the multiple-factor control process respective environment variable real data The data memory module is stored in, until recovering with the communication of the remote control terminal, switches to multiple-factor control process.
  10. A kind of 10. intelligent controller towards greenhouse, it is characterised in that including:Central processing unit and with the center Data memory module, communication module, analog-to-digital converting module and the relay module that processor communicates to connect respectively;
    The environmental data sensing of multiple environmental variance real data in the analog-to-digital converting module connection measurement target greenhouse The output end of device, the relay module connect the input of each environment adjustment actuating mechanism in the target greenhouse;
    The data memory module is used to store towards the intelligent control algorithm of greenhouse, environmental variance real data, control Implementation procedure data, multiple-factor control process respective environment variable real data, give and it is expected environmental data and control to fixed number According to;
    The central processing unit obtains each environmental variance real data by the analog-to-digital converting module, and is used for:
    Using single-factor control mode, the multiple environmental variance data are controlled respectively, and each environmental variance of corresponding record is actual Data and control implementation procedure data, until record number reaches setting demand, the single-factor control mode refer to respectively with Single environmental variance is control targe, individually controls each environmental variance data;
    Initial neural network model is built, and utilizes each environmental variance real data and the control implementation procedure data, The initial neural network model is trained, obtains control neural network model;
    And based on the multiple-factor control process respective environment variable real data and the given expectation environmental data, profit With the control neural network model, the control data-oriented of the multiple-factor control process is obtained, and passes through the relay The control data-oriented is issued to corresponding environment adjustment actuating mechanism by module, adjusts the respective environment in the target greenhouse Variable data.
CN201710923214.XA 2017-09-30 2017-09-30 It is a kind of towards the intelligent control method of greenhouse, system and controller Pending CN107678410A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201710923214.XA CN107678410A (en) 2017-09-30 2017-09-30 It is a kind of towards the intelligent control method of greenhouse, system and controller

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201710923214.XA CN107678410A (en) 2017-09-30 2017-09-30 It is a kind of towards the intelligent control method of greenhouse, system and controller

Publications (1)

Publication Number Publication Date
CN107678410A true CN107678410A (en) 2018-02-09

Family

ID=61139639

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201710923214.XA Pending CN107678410A (en) 2017-09-30 2017-09-30 It is a kind of towards the intelligent control method of greenhouse, system and controller

Country Status (1)

Country Link
CN (1) CN107678410A (en)

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108334139A (en) * 2018-03-01 2018-07-27 深圳春沐源控股有限公司 A kind of method and system of the greenhouse automation control based on big data
CN108762063A (en) * 2018-06-06 2018-11-06 吉林大学 A kind of heliogreenhouse wet curtain fan circulating ventilation cooling system and its control method
CN108781926A (en) * 2018-06-14 2018-11-13 淮安信息职业技术学院 Greenhouse irrigation system based on neural network prediction and method
CN109062296A (en) * 2018-08-27 2018-12-21 华夏易通国际物流有限公司 A kind of grain storage method based on temperature and humidity monitor
CN110400170A (en) * 2019-07-10 2019-11-01 北京耕智农业科技有限公司 A kind of method and device based on user feedback data adjustment crop supply
CN113093835A (en) * 2021-04-13 2021-07-09 广州中国科学院沈阳自动化研究所分所 Greenhouse control system and method based on edge controller

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103105246A (en) * 2012-12-31 2013-05-15 北京京鹏环球科技股份有限公司 Greenhouse environment forecasting feedback method of back propagation (BP) neural network based on improvement of genetic algorithm
CN103235620A (en) * 2013-04-19 2013-08-07 河北农业大学 Greenhouse environment intelligent control method based on global variable prediction model
CN203278890U (en) * 2013-04-17 2013-11-06 北京农学院 Linux-based remote greenhouse monitoring system
CN103605385A (en) * 2013-09-30 2014-02-26 中国农业大学 CO2 gas fertilizer fine regulation and control method and device used for solar greenhouse
CN106814782A (en) * 2017-01-19 2017-06-09 深圳前海弘稼科技有限公司 A kind of method and server for adjusting greenhouse

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103105246A (en) * 2012-12-31 2013-05-15 北京京鹏环球科技股份有限公司 Greenhouse environment forecasting feedback method of back propagation (BP) neural network based on improvement of genetic algorithm
CN203278890U (en) * 2013-04-17 2013-11-06 北京农学院 Linux-based remote greenhouse monitoring system
CN103235620A (en) * 2013-04-19 2013-08-07 河北农业大学 Greenhouse environment intelligent control method based on global variable prediction model
CN103605385A (en) * 2013-09-30 2014-02-26 中国农业大学 CO2 gas fertilizer fine regulation and control method and device used for solar greenhouse
CN106814782A (en) * 2017-01-19 2017-06-09 深圳前海弘稼科技有限公司 A kind of method and server for adjusting greenhouse

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
BAICHENG LIU ET AL.: "Research on Greenhouse Climate Control Based on RBF Neural Networks", 《PROCEEDINGS OF 2013 CHINESE INTELLIGENT AUTOMATION CONFERENCE》 *
刘鑫荣: "基于Internet和专家系统的温室环境监控系统的研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 *

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108334139A (en) * 2018-03-01 2018-07-27 深圳春沐源控股有限公司 A kind of method and system of the greenhouse automation control based on big data
CN108762063A (en) * 2018-06-06 2018-11-06 吉林大学 A kind of heliogreenhouse wet curtain fan circulating ventilation cooling system and its control method
CN108762063B (en) * 2018-06-06 2020-01-24 吉林大学 Circulating ventilation cooling system of wet curtain fan of sunlight greenhouse and control method thereof
CN108781926A (en) * 2018-06-14 2018-11-13 淮安信息职业技术学院 Greenhouse irrigation system based on neural network prediction and method
CN108781926B (en) * 2018-06-14 2021-10-22 江苏电子信息职业学院 Greenhouse irrigation system and method based on neural network prediction
CN109062296A (en) * 2018-08-27 2018-12-21 华夏易通国际物流有限公司 A kind of grain storage method based on temperature and humidity monitor
CN110400170A (en) * 2019-07-10 2019-11-01 北京耕智农业科技有限公司 A kind of method and device based on user feedback data adjustment crop supply
CN113093835A (en) * 2021-04-13 2021-07-09 广州中国科学院沈阳自动化研究所分所 Greenhouse control system and method based on edge controller

Similar Documents

Publication Publication Date Title
CN107678410A (en) It is a kind of towards the intelligent control method of greenhouse, system and controller
CN205161283U (en) Intelligence planter
CN105159366B (en) A kind of user behavior acquisition method for being suitable to greenhouse surroundings monitoring system
CN105897901A (en) Wholly artificial light type plant factory intelligent monitoring system based on Internet of Things (IOT)
CN204731617U (en) A kind of life cycle characteristic analysis system based on greenhouse gardening organic plant
CN104731135A (en) Control device and method for family farm
CN106873676A (en) A kind of wisdom agricultural data acquisition method and system based on Internet of Things
CN107608318A (en) A kind of intelligent home control system based on NB IOT
CN109976421A (en) A kind of intelligent Greenhouse control system and application method
CN203950191U (en) A kind of plant supervising device based on mobile device
CN105068586A (en) Temperature and humidity remote monitoring system suitable for multiple tobacco curing houses
CN103048987A (en) Internet-of-Things-based intelligent production monitoring terminal of agricultural greenhouse
CN203276003U (en) Silkworm chamber monitoring system based on Internet of Things
CN206115310U (en) Intelligent agriculture monitored control system based on thing networking
CN109952948A (en) A kind of remote visualization Intelligent irrigation system based on Internet of Things
CN108901474A (en) A kind of intelligence planting greenhouse and control method
CN107577266A (en) A kind of warmhouse booth intelligent controlling device
CN111366271A (en) White spirit solid state cellar intelligent temperature monitoring system based on bluetooth heterogeneous technology
Bai et al. The remote monitoring system of vegetable greenhouse
CN206671862U (en) Greenhouse environment factor intelligent controlling device based on PLC
CN206650698U (en) A kind of reading intelligent agriculture detection control apparatus based on internet
CN108064702A (en) A kind of pig house intelligence control system
CN209359164U (en) A kind of wireless automatic greenhouse wound membrane ventilating system
CN110716598A (en) Intelligent greenhouse control and management system based on Internet of things
CN208092577U (en) A kind of greenhouse remote monitoring system

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

Application publication date: 20180209

RJ01 Rejection of invention patent application after publication