CN102681438A - Crop greenhouse cultivation expert control system and crop disease diagnostic method - Google Patents
Crop greenhouse cultivation expert control system and crop disease diagnostic method Download PDFInfo
- Publication number
- CN102681438A CN102681438A CN2012101411153A CN201210141115A CN102681438A CN 102681438 A CN102681438 A CN 102681438A CN 2012101411153 A CN2012101411153 A CN 2012101411153A CN 201210141115 A CN201210141115 A CN 201210141115A CN 102681438 A CN102681438 A CN 102681438A
- Authority
- CN
- China
- Prior art keywords
- illness
- module
- decision
- crop
- user
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Landscapes
- Medical Treatment And Welfare Office Work (AREA)
Abstract
The invention belongs to the fields of facility agriculture, a computer technology and an automatic control technology. The invention provides a crop greenhouse cultivation expert control system and a disease diagnostic algorithm of a disease diagnosis subsystem thereof. The crop greenhouse cultivation expert control system is mainly characterized in that the crop greenhouse cultivation expert control system is connected with an environment parameter detecting system, a crop growth condition detecting system and an environment conditioning system, and has main functions of environment parameter decision of crop cultivation, growth condition classification, pathologic diagnosis indirection and the like. The disease diagnostic algorithm of the disease diagnosis subsystem comprises a crop disease characteristic knowledge encoding expression manner and a disease characteristic extraction algorithm. An expert system knowledge base can be replaced in an assorted manner according to relevance among subsystems of different crops, so as to be used as a greenhouse cultivation expert system for the different crops. The crop greenhouse cultivation expert control system provided by the invention has simple volume, is simple and easy to operate, can be independently used or used by means of being assorted with an external system, and other advantages. The crop greenhouse cultivation expert control system is very suitable for an agricultural greenhouse cultivation producer to use, and has a great development prospect.
Description
Technical field
The invention belongs to industrialized agriculture, computer technology and automatic control technology field.
Background technology
China is populous nation, and the agricultural development aspect is the main development field of China always, and along with continuous progress in science and technology, the installment agriculture growth field that automaticity is high becomes the importance of China's agricultural development.
The industrialized agriculture greenhouse production has robotization, intellectuality, characteristics that mechanization degree is high, and the regulator control system in the greenhouse can provide the crop growth environment of relative ideal, the quality and the security that improve agricultural product greatly simultaneously.
China region is wide, and the each department geographical conditions are different, and the development of industrialized agriculture greenhouse production can improve the growth conditions of crop-planting to a certain extent, prolongs the crops supply and demand time, improves agricultural output, is fit to the national conditions of China.
Currently be used for the crop greenhouse production and be mostly greenhouse culture environment monitoring or agriculture specialist system etc.; These systems provide the environmental management of greenhouse production; Perhaps the expertise of guiding agricultural production comprises variety classification, characteristic, the illness type of Different Crop and instructs the method for cultivating, and can carry out the monitoring and the regulation and control of greenhouse according to the suitable growth developing environment of crop; Information such as forecast maturation time, guides user is carried out the study of agricultural knowledge and the cultivation of crop.
But there are following problem in these agricultural cultivation expert systems and environmental monitoring system:
1. systemic-function is single not comprehensive
Agriculture specialist system or environmental monitoring system can only be single the functions such as control environment regulation and control of the expertise that agriculture aspect is provided or a certain crop; Though realized certain intellectuality and guiding aspect the suitable growth and development environment of crop and the arable farming creating, can not realize the comprehensive automation and the intelligent requirements of chamber crop cultivation.
2. limitation is strong, and the robotization of crop growth condition monitoring is not enough
The functions such as control environment regulation and administration that existing agricultural greenhouse cultivation system only provides to monocrop; When raise crop kind or kind change; Need carry out secondary development or software upgrading replacement to system, limitation is very big, and the growth and development state monitoring aspect of crop is seemed not enough; Can not judge the result and carry out handled of growing of crop automatically and reasonably, automaticity is not high.
Summary of the invention
The objective of the invention is to solve the problem that insufficiency face, intellectuality, robotization are low and limitation is strong of existing chamber cultivation system, the illness diagnosis algorithm of a kind of crop greenhouse production expert control system and illness diagnostic subsystem thereof is provided.
Provided by the invention is the crop greenhouse production expert control system of hardware platform with the embedded device; Be operation under embedded-type ARM 11 development systems and Windows CE 6.0 environment, comprise: Environmental Decision-making, growing state decision-making and illness are diagnosed three sub-systems;
The first, Environmental Decision-making subsystem comprises 4 modules:
Parameter monitoring module: comprise environmental parameter monitoring and user prompt.Parameter monitoring specifically comprises: temperature, humidity for the decision information that the real time environment parameter shows and the automatic decision module draws that the peripheral environmental monitoring system that connects monitors shows ... Show that etc. environmental parameter each environmental parameter result of decision shows.The warning function of user prompt for when the serious unsuitable crop of current environment grows, sending to the user.
Manually regulate and control module: comprise the control interface that the user regulates and control environment conditioning equipment manually.This module links to each other with the peripheral environment regulator control system, can be through the regulation and control button that is provided with in the module be carried out the control of peripheral environment adjusting device, as: temperature regulator, the control of equipment such as humidistat.Specifically comprise: temperature, humidity ... Regulate the data presentation of button and current detection value and regulated value Deng the environment conditioning equipment that connects.
Automatic decision module: for the Environmental Decision-making subsystem environmental parameter is made a strategic decision, be the running background module.Comprised corresponding environmental parameter decision making algorithm in this module, for the Environmental Decision-making subsystem provides the environmental parameter decision making function and when environmental parameter does not seriously suit crop growth, sends signal to the parameter monitoring module.
Data base administration calling module: comprise database storing management and KBM.Database storing is managed the real-time storage of the environmental parameter that has realized the collection of peripheral environment monitoring system and is called; Manually regulate and control module and the automatic decision module provides real time data to the parameter monitoring module; And database management function is provided to the user, comprise interpolation, deletion, the query function of data.KBM comprises interpolation, deletion, renewal, the query function of knowledge in the knowledge base; And can carry out calling of different knowledge bases of crop different bearing stage according to the growing stage signal of growing state decision-making subsystem, this calls to automatic decision module and parameter monitoring and opens with manual regulation and control module.
Its method of operation is following:
The peripheral environment monitoring system is imported the environmental parameter that collects into and is supplied Environmental Decision-making system automatic decision module invokes in the database; This module is carried out the reasoning decision-making of environmental parameter according to the expertise in the knowledge base; And the result of decision is transferred to the environment conditioning system, the demonstration of the real-time parameter and the result of decision is shown by the parameter monitoring module; Manually the regulation and control module has the function that the environment conditioning system is manually controlled; Data base call administration module and the database and the knowledge base of Environmental Decision-making system managed.
The second, growing state decision-making subsystem comprises 3 modules:
User prompt module: comprise the demonstration of the current growing state of crop.Wherein growing state comprises: whether plant physiology puberty, crop the result of normal development.The testing result of user prompt module receiving and analyzing decision-making module crop growth stage decision information and peripheral growing state detection system also shows.
Analysis decision module: the automatic arbitration functions of plant physiology developmental stage is provided, is the running background module.The judgement computational algorithm that comprises the plant physiology developmental stage in the module carries out the crop growth judgement decision-making in period through the parameter that reads the environment decision-making subsystem.
Data base administration calling module: comprise database storing management and KBM.The database storing management has realized the real-time storage of the detection data that peripheral growing state detection system draws and has called, and to user prompt module real time data is provided, and to the user database management function is provided, and comprises interpolation, deletion, the query function of data.KBM comprises interpolation, deletion, renewal, the query function of knowledge in the knowledge base; And can carry out calling of different knowledge bases of crop different bearing stage according to the growing stage signal of growing state decision-making subsystem, this calls to user prompt module and analysis decision module and opens.
Its method of operation is following:
System carries out the judgement of crop developmental stage through the calculating of physiological development time, and judged result is transferred to the knowledge base that Environmental Decision-making system and illness diagnostic system carry out the crop different developmental phases with aspect calls; This system links to each other with peripheral crop growth conditions detection system, and through the judgement of crop growth conditions detection system result being carried out signal output, this signal is transferred to the illness diagnostic system to carry out the reasoning of illness characteristic and call, for the user provides guidance.
Three, illness diagnostic subsystem comprises 3 modules:
User prompt module: comprise manually-operated and user prompt.Manually-operated: system interface provides the illness that possibly exist when preceding crop mark sheet; The user can carry out the illness Feature Selection according to the characteristic of crop in the illness mark sheet; Choosing the result can show in the Feature Selection hurdle, and carries out illness feature coding and search for the illness reasoning module Feature Selection result transmission, when the user clicks diagnostic button; The illness reasoning module carries out the diagnosis of illness reasoning, will be transferred to user prompt module simultaneously and carry out result's demonstration.User prompt: provide the result of automatic illness reasoning to show and manually-operated operation helping prompt.
Illness reasoning module: realized illness diagnostic reasoning function, be the running background module.Wherein comprise illness feature coding mechanism and illness feature extraction algorithm and inference machine.Illness feature coding mechanism is encoded the characteristic that the user selects; Inference machine carries out reverse search and reasoning based on feature coding; If the user does not select to diagnose; Then reverse search goes out with the user to select under the characteristic the compound illness characteristic of illness to upgrade the illness mark sheet, if the user selects to diagnose then to release the illness result.The illness feature extraction algorithm can calculate the illness performance degree of membership of accordingly result according to the testing result of peripheral upgrowth situation detection system, and the maximum feature extraction as illness of degree of membership is come out, and characteristic is transferred to inference machine carries out reasoning.
Data base administration calling module: comprise database storing management and KBM.The database storing management has realized the real-time storage of the detection data that peripheral growing state detection system draws and has called; The result of automatic illness reasoning is provided to user prompt module; And database management function is provided to the user, comprise interpolation, deletion, the query function of data.KBM comprises interpolation, deletion, renewal, the query function of knowledge in the knowledge base, and can carry out calling of different knowledge bases of crop different bearing stage according to the growing stage signal of growing state decision-making subsystem, and this calls to the illness reasoning module and opens.
Its concrete method of operation is following:
System is divided into two kinds of diagnostic modes, can carry out simultaneously.Carry out the automatic diagnosis and artificial input diagnosis of illness according to user's request.
The automatic diagnosis mode: reasoning and knowledge base that signal through accepting the growing state decision system and upgrowth situation detection system signal carry out illness are automatically called.
Artificial input diagnostic mode:, carry out the reasoning decision-making through the selection input of user to the illness characteristic.The illness characteristic that the user is provided gathers, and draws illness title, characteristic and processing mode, and guidance is provided.
Wherein the user prompt module of different sub-systems is different, and the data base administration calling module is also different.Expert control system of the present invention links to each other with peripheral environment monitoring system, upgrowth situation detection system and environment conditioning system simultaneously, has the function of environment conditioning decision-making, growing state detection and illness diagnosis.
Described diagnostic method comprises crop pest diagnosis coding knowledge representation mode and corresponding inference mechanism thereof, and this method can adapt to the replacing and the illness diagnosis needs of Different Crop knowledge base, and concrete diagnosis algorithm is following:
The illness feature extraction algorithm; The notion of the Imperfect Information Systems in this algorithm use rough set theory is summed up incomplete decision information table, proposes illness characteristic attribute rough set; And to have the support collection class of identical description; Promptly the description collection is basic module, merges to constitute similar collection, calculates the algorithm of illness characteristic blur level with the derivation ambiguity function; The data obfuscation that the upgrowth situation detection system is gathered forms the incomplete decision table of gathering sample and illness characteristic, and calculates that corresponding membership function value reaches that feature identification is chosen and then the purpose diagnosed of illness automatically.
Hardware characteristics of the present invention is:
The master control expert system is moved on ARM11, adopts the S3C6410 processor as core controller.
The present invention is a system environments with Windows CE 6.0, has developed with respective environment parameter detecting, crop growth conditions detection and environment conditioning system to be associated, and is suitable for the master control expert system of greenhouse production.
The present invention links to each other with peripheral environment monitoring system, upgrowth situation detection system and environment conditioning system, and environmental monitoring system is transferred to the Environmental Decision-making subsystem through the Zigbee wireless transport module with the real time environment parameter that collects and carries out the environmental parameter decision-making; The upgrowth situation detection system through the CAN bus testing result is transferred to the growing state decision system and the judgement and the illness diagnosis in crop growth stage are carried out in the illness diagnosis.
Advantage of the present invention and beneficial effect:
This system's principal character is; Link to each other with supporting environmental parameter detection system, crop growth conditions detection system and environment conditioning system and can constitute crop greenhouse production control system; Have major functions such as environmental parameter decision-making, growing state classification and pathological diagnosis guidances; And corresponding Data Receiving and management function are provided, its software knowledge base can carry out complete replacing according to different crops.
The present invention has that volume is little, simple and easy, the function of operation is strong, can use separately or advantage such as external system support use.The present invention is the combination of information hi-tech, automatic control technology and conventional agriculture knowledge; The integrated relevant agricultural knowledge of crop greenhouse production; According to the judgement of classifying of the testing result of crop growth conditions detection system; Data according to the collection of environmental parameter detection system are the automatic adjusting that environment conditioning provides the result of decision to carry out environment, and for the illness situation of crop the guidance of individual features information and disposal route are provided.Be beneficial to management and the problem that agriculture personnel carry out greenhouse production and instruct mode, meet the developing direction of present industrialized agriculture, solve the incomplete problem of agriculture personnel specialty knowledge, help improving agricultural output, have huge development prospect.
Description of drawings
Fig. 1 is the whole software module structure synoptic diagram of the present invention.
Fig. 2 is an overall hardware configuration synoptic diagram of the present invention.
Fig. 3 is the tree structure synoptic diagram of illness diagnosis coding knowledge representation mode of the present invention.
Fig. 4 is that illness of the present invention is diagnosed forward and reverse search inference mode synoptic diagram.
Fig. 5 is an illness diagnostic reasoning program flow diagram of the present invention.
Embodiment
Environmental parameter detection system of the present invention is gathered environmental parameter in real time, compiles all collection result through the transmission of Zigbee wireless transport module, imports expert system into by serial ports.Plant growth situation detection system is carried out the collection of image through camera; By the CAN bus result is transmitted into expert system after treatment; The result of decision of expert system is passed through screen displaying, and is transferred to the auto-control that the environment conditioning system carries out environment.
The present invention has three sub-systems:
The first, Environmental Decision-making subsystem comprises 4 modules:
Parameter monitoring module: comprise environmental parameter monitoring and user prompt.Parameter monitoring specifically comprises: temperature, humidity for the decision information that the real time environment parameter shows and the automatic decision module draws that the peripheral environmental monitoring system that connects monitors shows ... Show that etc. environmental parameter each environmental parameter result of decision shows.The warning function of user prompt for when the serious unsuitable crop of current environment grows, sending to the user.
Manually regulate and control module: comprise the control interface that the user regulates and control environment conditioning equipment manually.This module links to each other with the peripheral environment regulator control system, can be through the regulation and control button that is provided with in the module be carried out the control of peripheral environment adjusting device, as: temperature regulator, the control of equipment such as humidistat.Specifically comprise: temperature, humidity ... Regulate the data presentation of button and current detection value and regulated value Deng the environment conditioning equipment that connects.
Automatic decision module: for the Environmental Decision-making subsystem environmental parameter is made a strategic decision, be the running background module.Comprised corresponding environmental parameter decision making algorithm in this module, for the Environmental Decision-making subsystem provides the environmental parameter decision making function and when environmental parameter does not seriously suit crop growth, sends signal to the parameter monitoring module.
Data base administration calling module: comprise database storing management and KBM.Database storing is managed the real-time storage of the environmental parameter that has realized the collection of peripheral environment monitoring system and is called; Manually regulate and control module and the automatic decision module provides real time data to the parameter monitoring module; And database management function is provided to the user, comprise interpolation, deletion, the query function of data.KBM comprises interpolation, deletion, renewal, the query function of knowledge in the knowledge base; And can carry out calling of different knowledge bases of crop different bearing stage according to the growing stage signal of growing state decision-making subsystem, this calls to automatic decision module and parameter monitoring and opens with manual regulation and control module.
Its method of operation is following:
The peripheral environment monitoring system is imported the environmental parameter that collects into and is supplied Environmental Decision-making system automatic decision module invokes in the database; This module is carried out the reasoning decision-making of environmental parameter according to the expertise in the knowledge base; And the result of decision is transferred to the environment conditioning system, the demonstration of the real-time parameter and the result of decision is shown by the parameter monitoring module; Manually the regulation and control module has the function that the environment conditioning system is manually controlled; Data base call administration module and the database and the knowledge base of Environmental Decision-making system managed.
The second, growing state decision-making subsystem comprises 3 modules:
User prompt module: comprise the demonstration of the current growing state of crop.Wherein growing state comprises: whether plant physiology puberty, crop the result of normal development.The testing result of user prompt module receiving and analyzing decision-making module crop growth stage decision information and peripheral growing state detection system also shows.
Analysis decision module: the automatic arbitration functions of plant physiology developmental stage is provided, is the running background module.The judgement computational algorithm that comprises the plant physiology developmental stage in the module carries out the crop growth judgement decision-making in period through the parameter that reads the environment decision-making subsystem.
Data base administration calling module: comprise database storing management and KBM.The database storing management has realized the real-time storage of the detection data that peripheral growing state detection system draws and has called, and to user prompt module real time data is provided, and to the user database management function is provided, and comprises interpolation, deletion, the query function of data.KBM comprises interpolation, deletion, renewal, the query function of knowledge in the knowledge base; And can carry out calling of different knowledge bases of crop different bearing stage according to the growing stage signal of growing state decision-making subsystem, this calls to user prompt module and analysis decision module and opens.
Its method of operation is following:
System carries out the judgement of crop developmental stage through the calculating of physiological development time, and judged result is transferred to the knowledge base that Environmental Decision-making system and illness diagnostic system carry out the crop different developmental phases with aspect calls; This system links to each other with peripheral crop growth conditions detection system, and through the judgement of crop growth conditions detection system result being carried out signal output, this signal is transferred to the illness diagnostic system to carry out the reasoning of illness characteristic and call, for the user provides guidance.
Three, illness diagnostic subsystem comprises 3 modules:
User prompt module: comprise manually-operated and user prompt.Manually-operated: system interface provides the illness that possibly exist when preceding crop mark sheet; The user can carry out the illness Feature Selection according to the characteristic of crop in the illness mark sheet; Choosing the result can show in the Feature Selection hurdle, and carries out illness feature coding and search for the illness reasoning module Feature Selection result transmission, when the user clicks diagnostic button; The illness reasoning module carries out the diagnosis of illness reasoning, will be transferred to user prompt module simultaneously and carry out result's demonstration.User prompt: provide the result of automatic illness reasoning to show and manually-operated operation helping prompt.
Illness reasoning module: realized illness diagnostic reasoning function, be the running background module.Wherein comprise illness feature coding mechanism and illness feature extraction algorithm and inference machine.Illness feature coding mechanism is encoded the characteristic that the user selects; Inference machine carries out reverse search and reasoning based on feature coding; If the user does not select to diagnose; Then reverse search goes out with the user to select under the characteristic the compound illness characteristic of illness to upgrade the illness mark sheet, if the user selects to diagnose then to release the illness result.The illness feature extraction algorithm can calculate the illness performance degree of membership of accordingly result according to the testing result of peripheral upgrowth situation detection system, and the maximum feature extraction as illness of degree of membership is come out, and characteristic is transferred to inference machine carries out reasoning.
Data base administration calling module: comprise database storing management and KBM.The database storing management has realized the real-time storage of the detection data that peripheral growing state detection system draws and has called; The result of automatic illness reasoning is provided to user prompt module; And database management function is provided to the user, comprise interpolation, deletion, the query function of data.KBM comprises interpolation, deletion, renewal, the query function of knowledge in the knowledge base, and can carry out calling of different knowledge bases of crop different bearing stage according to the growing stage signal of growing state decision-making subsystem, and this calls to the illness reasoning module and opens.
Its concrete method of operation is following:
System is divided into two kinds of diagnostic modes, can carry out simultaneously.Carry out the automatic diagnosis and artificial input diagnosis of illness according to user's request.
The automatic diagnosis mode: reasoning and knowledge base that signal through accepting the growing state decision system and upgrowth situation detection system signal carry out illness are automatically called.
Artificial input diagnostic mode:, carry out the reasoning decision-making through the selection input of user to the illness characteristic.The illness characteristic that the user is provided gathers, and draws illness title, characteristic and processing mode, and guidance is provided.
Wherein the user prompt module of different sub-systems is different, and the data base administration calling module is also different.Expert control system of the present invention links to each other with peripheral environment monitoring system, upgrowth situation detection system and environment conditioning system simultaneously, has the function of environment conditioning decision-making, growing state detection and illness diagnosis.
The Environmental Decision-making system carries out the decision-making of environmental parameter and classification that output, growing state decision system are carried out growing state is judged and the signal of corresponding expert knowledge library calls transmission, the illness diagnostic system provides illness diagnosis guidance.
Its detailed process is following:
1. plant growth situation detection system will detect in the growing state decision system that result is transferred to expert system; The classification that this subsystem carries out the plant growth situation according to the information that receives judges, the expertise that decision is selected for use also is transferred to Environmental Decision-making system and illness diagnostic system with the respective calls signal and carries out expertise and call.
2. after the Environmental Decision-making system receives the expertise call signal of growing state decision system, call expertise according to signal.Environmental parameter in the greenhouse of reception environment parameter detecting system transmission; And carry out the decision-making of environmental parameter according to expertise; Then the result of decision is transferred to the environment conditioning system and carries out environment conditioning, Environmental Decision-making result can be shown by the user prompt module of this subsystem.Wherein the Environmental Decision-making system has the automatic module that converts manual adjustments into of regulating to the environment conditioning system.
3. plant growth situation detection system has illness class judgement, can call the guidance that expertise carries out the illness disposal route automatically behind the signal of illness diagnostic system reception growing state decision system.The illness diagnostic subsystem also can carry out the purpose that instructs that calling of illness knowledge reaches greenhouse production through user's operation, and its concrete diagnostic procedure is following:
Artificial input diagnosis:
Provide in the system interface when preceding crop illness characteristic hurdle; Comprising the illness characteristic of all illnesss of crop, the user can carry out the interpolation of choosing of illness characteristic according to this characteristic hurdle, selects to add the result and will show and be transferred to inference machine simultaneously; Inference machine carries out reverse search according to selected illness characteristic; Obtain the illness characteristic relevant, and upgrade illness characteristic hurdle to dwindle the hunting zone, when the user selects to carry out the illness diagnosis with selected characteristic; Inference machine is encoded selected illness feature illness feature coding mode, infers diagnostic result and demonstration then.
Automatic diagnosis:
Peripheral upgrowth situation detection system is given the illness diagnostic subsystem with detected result transmission, and inference machine is categorized as the testing result degree: normal, light and heavy.Carry out the decision attribute degree of membership according to the illness feature extraction algorithm and whether calculate conduct with the foundation of this result as the illness characteristic; Reach the purpose of illness feature extraction, after obtaining the illness characteristic, then can carry out feature coding and carry out the illness diagnosis then automatically according to encoding mechanism.
Expert system is stored in the embedded device and links to each other with peripheral system, and its connection data transmission relations are:
The environmental parameter detection system is imported the data of gathering in the real-time data base of environmental parameter decision system into, for this system call through the Zigbee wireless transmission; The crop growth conditions detection system will be handled through the CAN bus and the result imported in the real-time data base of growing state system for calling; The automatic decision module of Environmental Decision-making system is transferred to the environment conditioning system with the result of decision.
The practical implementation algorithmic descriptions is following:
1. growing state decision system of the present invention is judged sorting algorithm to the plant growing state:
Adopt EMS to carry out the growth and development stage evaluation algorithm:
At first the environmental parameter detection system is imported detected environmental parameter in the database into and is stored, and its storage format is: date (xx branch during xxxx xx month xx day xx), environmental parameter (temperature, humidity, carbon dioxide content ...);
The account form that its growth and development stage is judged is the physiological development Time Calculation, after calculating the physiological development time, carries out the stage of development and judges;
Its arthmetic statement is:
Adopt the upgrowth situation detection system to carry out the growth and development stage evaluation algorithm:
2. exact algorithm is adopted in the decision-making of Environmental Decision-making system environments parameter of the present invention, and crop requires parameter value certain at its optimum to environment of certain growth period, so Parameter Decision Making adopts accurate reasoning.
Because parameter is many; Consider the rate of change problem of regulation and control cost problem and parameter, some parameter is difficult for regulating, and some parameter needs real-time update; So in order to reduce the expense of system's double counting; System carries out parameter acquisition and changes judgement, according to judging whether the information decision calculates decision-making, and it specifically describes as follows:
3 crop illness diagnosis algorithms:
The illness diagnosis algorithm of crop greenhouse production expert control system illness diagnostic subsystem comprises crop pest diagnosis coding knowledge representation mode and corresponding inference mechanism thereof; It is characterized in that to adapt to the replacing and the illness diagnosis needs of Different Crop knowledge base, make expert system not need overlapping development.It is described as follows:
Crop is different in the illness characteristic of different times; Therefore the illness characteristic can be summarized according to the developmental stage of crop; Like the tomato crop at the available (S1 of the gray mold symptom of seedling phase (be on the blade V-shape black scab, grey mold go out the grey hair at moist duration, the fruit face shows as water soaking mode and rots); S2, S3) expression, gray mold is represented with H.The rest may be inferred, and in a stage of development, all illness characteristics are with array S (S1, S2; S3 ... Si) expression, disease is used alphabetical A, B; C...... expression, in order to know the relation between expression characteristic array and the disease, available tree structure figure representes, like Fig. 3.
Can clearly obtain the corresponding relation between characteristic array S and illness A, the B...... by Fig. 3, (S8 S9) has represented illness A for S1, S2, and (S1 S3) has represented illness C to array, array (S13) expression illness G etc. like array.The common characteristic that is characterized as several kinds of diseases of root node among Fig. 3, therefore, a kind of disease of feature instantiation is many more, and the node location at its place is the closer to root node so.Last position is partly represented a kind of disease title among Fig. 3, and the characteristics of incidence of this disease is an illness characteristic all on the branch of front, when the illness characteristic summation on the branch can accurately clearly be come out disease title correspondence together the time.Be that S1, S2, S8, the S9 crop when falling ill for illness A shows characteristic.
When clearly sorting table illustrates the corresponding relation of illness characteristic and disease from above-mentioned tree structure, and need reasonably put in order, make it become the computerese that can move its corresponding relation.For ease the characteristic array is encoded, represent that with digital quantity i (the characteristic array like illness A can use (1,2,8 for the array of illness characteristic; 9)) replace characteristic Si to encode, coded system is the sequencing according to node, carries out sorting coding according to digital quantity; Characteristic (S1, S2, S8, the S9) digital quantity that is illness A has 1,2,8; 9, coded system is string codeA=' 1 '+' 2 '+' 8 '+' 9 ' 1289, string codeC=' 1 '+' 3 '=13.After adopting above-mentioned coded system, the illness coding that obtains can be mapped with the illness characteristic easily.
Coded system go replication problem:
Based on above-mentioned coded system, after selected characteristic is encoded, can express a kind of disease accurately.But same coding also can obtain string codeG=13, and the coding string codeC=13 repetition with illness C can cause The reasoning results inaccurate like this.Therefore, this paper has designed a kind of coded system of repetition of going and has addressed this problem.
As (i < 100) when the root node independent is selected, coding rear padding ' 0 ', string codeC=103 like this, string codeG=13 has avoided the coding replication problem.This principle of repeated encoding mode of going is that after the benefit " 0 ", the result of the non-independent feature coding that is carried out need exceed the illness characteristic maximal value of coding separately.
Go the service condition of repeated encoding mode:
Go repetitive mode to find out by above-mentioned coding, the illness characteristic array that digital quantity i representes, the span of i is conditional on the replication problem going; When (i 10) time; Go the repeated encoding mode promptly not have necessity of using, because string is codeC=13, and S13 < did not exist at i in 10 o'clock.In like manner, when 100 <i < 1000, make the repeated encoding mode that spends; Coding string codeC=103; But because the variation of i span disease X possibly occur so, its illness is characterized as independent No. 103 illness characteristics (being S103); Coding string codeX=103 goes the repeated encoding mode can produce mistake at this moment.In order to address this is that, according to the scope of illness characteristic with go the principle of repeated encoding mode, when i < 10 the time, need not use this method; When 10 <i < 100, use the independent feature coding to mend the mode of " 0 "; When 100 <i < 1000; On principle, then need the result of independent coding be exceeded 1000, because independent feature coding minimum code condition is two kinds of characteristics; Then can get double-digit coding result behind the coding; In order to make it exceed 1000 scope, need complement code " 00 " so, the rest may be inferred solves repeated encoded question.
Certainly, go also can adopt on the problem of repetition other symbol or digital form to go to replace " 0 " in complement code, its principle is: can be received the complement code mode that does not influence its repeated encoding property by the expert system programming mode.
According to this coded system with characteristic and disease at once, but inference machine carries out the coupling of illness with regard to direct coding, fast and effeciently carries out calling of reasoning and knowledge.
The algorithm that the inference machine of illness diagnosis adopts is forward and reverse searching algorithm; After adopting above-mentioned numbering scheme to carry out knowledge representation and being stored in the knowledge base; The user can choose according to the crop characteristic that the illness diagnostic module is enumerated out, and inference machine is encoded according to the characteristic of selected taking-up, and forward lookup; The possible illness of selected characteristic is accessed from knowledge base, leave the intermediate storage space in.At this moment, diagnose, then the result is contingent all illnesss of selected characteristic.If further select; Inference machine recompile then; And carry out reverse search, with selected characteristic the institute of the illness that possibly represent might feature selecting come out, and reduce the scope and carry out the renewal of illness feature list; Next step that guides thus that the user carries out characteristic chosen, and finally carries out the illness diagnosis until the user.Its working method such as Fig. 4, program flow diagram such as Fig. 5.
The illness diagnosis algorithm of crop greenhouse production expert control system illness diagnostic subsystem adopts the notion of the Imperfect Information Systems in the rough set theory to carry out crop illness feature extraction diagnosis; It is characterized in that summing up incomplete decision information table; Constitute the rough set of illness characteristic, and with support collection class (describing collection) with identical description as basic module, merge and constitute similar collection; With the algorithm of the illness of ambiguity function calculating everywhere characteristic blur level, its arthmetic statement is following:
Key concept:
If (X ST) is an infosystem to U=, and wherein X is the limited domain of non-NULL, and the element among the X is called object, and ST is limited non-NULL property set, for each s ∈ S, s:X → V
s(V
sBe different from the civilian Vs in back) and
S (x) ∈ V ∞ claim V
sCodomain for attribute s.If some property value in the infosystem is default or part knows that such system is called Imperfect Information Systems, property value function s (x) can be defined as from X to V
sCollection value mapping of power set.
Incomplete decision table U=(X, ST ∪ { d}), X={x1 wherein, x2; X3 ... }, ST={s1, s2......} represent the condition flag attribute of disease characteristic; D represents the decision-making characteristic attribute of disease characteristic, Vs={Vs1, and Vs2 ... represent the index of characteristic.
In the algorithmic notation with the ∧ presentation logic " with ".If v ∈ Vs ∈ S, attribute is to (s, v) being called is the unit basically of S.The logical ∧ of the basic unit of all S or it is connected to become S and describes, and establishes t for describing, if (the s of unit basically; V) be present among the t; Claim (s, v) ∈ t is if
(s; V) ∈ t claims that then t is the full description of S.‖ t ‖={ x ∈ X:v ∈ s (x); (s, v) ∈ t} is called the support collection of t.If t and a are that two descriptions and S (t) ∩ S (a) are empty sets, can obtain ‖ t ∧ a ‖=‖ t ‖ ∩ ‖ a ‖.
If DES (S)={ t:t is the description of S to
note; ‖ t ‖ ≠ Φ }.
For t ∈ DES (S) arbitrarily, if S (t)=S, then t is the description of S completely, note FDES (S)={ t:t is the description fully of S }.If U={X; ST} is an Imperfect Information Systems; If for s ∈ ST and v=Vs v ∈ s (x) ∩ s (y) arbitrarily; Promptly
x and y are called as about attribute s similar; Equally; if
t ∈ FDES (S), x, y ∈ X; X then; Y is called as about similar and if only if
these similarity relations of S and just has been divided into limited domain X as basic module several basic modules of (call and describe collection), and these describe the covering that the collection class has constituted X, are designated as X/S={ ‖ t ‖: t ∈ FDES (S) }; The description collection class that comprises x is designated as S (x), i.e. S (x)={ ‖ t ‖: t ∈ FDES (S) }
The rough set structure:
Definition is established
With
XI about S down approximate and go up approximate be designated as respectively S (XI) and
Wherein
S (XI)-
Be called the border of XI, be designated as BN about S
s(XI).Annotate: defined here being similar to down and upward approximate set, element teacher wherein describes the collection class, rather than the sub-set of X.
With set X in element x have the set that the element of identical description constitutes and call similar collection, be designated as S
s(x), S then
s(x)=∪ ‖ t ‖: x ∈ ‖ t ||, t ∈ FDES (S) }
Ambiguity function is defined as:
S
s(u) be called about property set S and belong to the degree of membership of gathering Y, meaning directly perceived is that u belongs to the blur level of gathering Y.
Carry out algorithmic descriptions below for example:
Be provided with incomplete decision table U, U=(X, ST ∪ { d}), X={x1 wherein, x2 about crop sample and illness characteristic; X3 ... x7} represents different sample sets, and ST={s1, s2} represent the condition flag of disease characteristic, and d represents the decision-making characteristic (promptly judging a kind of principal character basis for estimation of illness performance) of this illness; Vs={Vs1, Vs2} represent the index of characteristic, like leaf color Vs={L, N; H}, L represent the color jaundice, and N representes that color is normal, and H representes that color is dark excessively.Vs1={L is arranged, N, H}.Attribute list such as table 1:
| X | S1 | S2 | d |
| x1 | N | N | N |
| x2 | N、H | N | H |
| x3 | N | N、H | N |
| x4 | L | L | L |
| x5 | H | N、H | H |
| x6 | N、H | H | H |
| x7 | N、L | N | L |
The incomplete decision table of table 1
Then according to the make of above-mentioned table, notion and rough set:
X2, x5, x6, the value about attribute S1 all have " H ", then they should belong to same description tired ‖ (a, H) ‖, i.e. ‖ (a, H) ‖={ x2, x5, x6} about attribute S1.Use S1, S2, ST={S1, the description collection class of S2} can be represented to classify:
X/{S1}={‖（S1，N）‖，‖（S1，H）‖，‖（S1，L）‖}={{x1，x2，x3，x6，x7}，{x2，x5，x6}，{x4，x7}}。
X/{S2}={‖（S2，N）‖，‖（S2，H）‖，‖（S2，L）‖}={{x1，x2，x3，x5，x7}，{x3，x5，x6}，{x4}}。
X/ST={‖（S1，N）∧（S2，N）‖，‖（S1，N）∧（S2，H）‖，‖（S1，H）∧（S2，N）‖，‖（S1，H）∧（S2，H）‖，‖（S1，L）∧（S2，L）‖，‖（S1，L）∧（S2，N）‖}={{x1，x2，x3，x7}，{x3，x6}，{x2，x5}，{x5，x6}，{x4}，{x7}}。
Also see easily simultaneously: { S1} (x2)={ { x1, x2, x3, x6, x7}, { x2, x5, x6}}.
If XI={x2, x5, x6},
Can get through calculating: S1} (XI)=x2, x5, x6}},
{S2}(XI)=Φ，
ST(XI)={{x5，x6}}，
BN
s(XI)={{x1，x2，x3，x7}，{x3，x6}}。
Similar collection: S
s(x1)={ x1, x2, x3, x7}, S
s(x2)={ x1, x2, x3, x5, x7}, S
s(x3)={ x1, x2, x3, x6, x7}, S
s(x4)={ x4}, S
s(x5)={ x2, x5, x6}, S
s(x6)={ x3, x4}, S
s(x7)={ x1, x2, x3, x7}.
{N}={x∈X：d（x）＝N}={x1,x3},{H}={x∈X：d（x）=H}={x2,x5,x6},{L}={x∈X：d（x）=L}={x4，x7}。
Then { N}, { H}, { L} has constituted the division of X, and { N}, { H}, { blur level of L} constitutes coarse membership function table, like table 2 in calculating according to ambiguity function
| X | N | | L |
| x1 | |||
| 1/2 | 1/4 | 1/4 | |
| |
2/5 | 2/5 | 1/5 |
| |
2/5 | 2/5 | 1/5 |
| x4 | 0 | 0 | 1 |
| x5 | 0 | 1 | 0 |
| |
1/2 | 1/2 | 0 |
| |
1/2 | 1/4 | 1/4 |
The coarse membership function table of table 2
So just can the degree of membership value from table 2 be clear that whether each sample elements is normal, its criterion is the functional value of each sample characteristics index.When the functional value of characteristic performance very big, just can confirm that there is corresponding illness characteristic in this sample crop, reach the purpose of crop illness feature extraction, thereby carry out the illness diagnostic reasoning.
Claims (2)
1. one kind is the crop greenhouse production expert control system of hardware platform with the embedded device; It is characterized in that this expert control system moves under embedded-type ARM 11 development systems and Windows CE 6.0 environment, comprising: Environmental Decision-making, growing state decision-making and illness are diagnosed three sub-systems;
The first, Environmental Decision-making subsystem comprises 4 modules:
Parameter monitoring module: comprise environmental parameter monitoring and user prompt; Parameter monitoring specifically comprises for the decision information that the real time environment parameter shows and the automatic decision module draws that the peripheral environmental monitoring system that connects monitors shows: environmental parameter shows that each environmental parameter result of decision shows; User prompt is the warning function that the serious suitable crop of current environment sends to the user when growing;
Manually regulate and control module: comprise the control interface that the user regulates and control environment conditioning equipment manually; This module links to each other with the peripheral environment regulator control system, carries out the control of the adjusting device in the peripheral environment regulation and control regulator control system through the regulation and control button that is provided with in the module; The control interface comprises the data presentation of environment conditioning equipment adjusting button and current detection value and regulated value;
Automatic decision module: for the Environmental Decision-making subsystem environmental parameter is made a strategic decision, be the running background module; Comprised corresponding environmental parameter decision making algorithm in this module, for the Environmental Decision-making subsystem provides the environmental parameter decision making function and when environmental parameter does not seriously suit crop growth, sends signal to the parameter monitoring module;
Data base administration calling module: comprise database storing management and KBM; Database storing is managed the real-time storage of the environmental parameter that has realized the collection of peripheral environment monitoring system and is called; To the parameter monitoring module, manually regulate and control module and the automatic decision module provides real time data; And database management function is provided to the user, comprise interpolation, deletion, the query function of data; KBM comprises interpolation, deletion, renewal, the query function of knowledge in the knowledge base; And carrying out calling of different knowledge bases of crop different bearing stage according to the growing stage signal of growing state decision-making subsystem, this calls to automatic decision module, parameter monitoring module and manually regulates and control module and open;
The second, growing state decision-making subsystem comprises 3 modules:
User prompt module: comprise the demonstration of the current growing state of crop; Wherein growing state comprises: whether plant physiology puberty, crop the result of normal development; The testing result of user prompt module receiving and analyzing decision-making module crop growth stage decision information and peripheral growing state detection system also shows;
Analysis decision module: the automatic arbitration functions of plant physiology developmental stage is provided, is the running background module; The judgement computational algorithm that comprises the plant physiology developmental stage in the module carries out the crop growth judgement decision-making in period through the parameter that reads the environment decision-making subsystem;
Data base administration calling module: comprise database storing management and KBM; The database storing management has realized the real-time storage of the detection data that peripheral growing state detection system draws and has called, and to user prompt module real time data is provided, and to the user database management function is provided, and comprises interpolation, deletion, the query function of data; KBM comprises interpolation, deletion, renewal, the query function of knowledge in the knowledge base; And carrying out calling of different knowledge bases of crop different bearing stage according to the growing stage signal of growing state decision-making subsystem, this calls to user prompt module and analysis decision module and opens;
Three, illness diagnostic subsystem comprises 3 modules:
User prompt module: comprise manually-operated and user prompt; Manually-operated: system interface provides the illness that possibly exist when preceding crop mark sheet; The user carries out the illness Feature Selection according to the characteristic of crop in the illness mark sheet; Choosing the result can show in the Feature Selection hurdle, and carries out illness feature coding and search for the illness reasoning module Feature Selection result transmission, when the user clicks diagnostic button; The illness reasoning module carries out the diagnosis of illness reasoning, simultaneously diagnostic result is transferred to user prompt module and carries out result's demonstration; User prompt: provide the result of automatic illness reasoning to show and manually-operated operation helping prompt;
Illness reasoning module: realized illness diagnostic reasoning function, be the running background module; Wherein comprise illness feature coding mechanism and illness feature extraction algorithm and inference machine; Illness feature coding mechanism is encoded the characteristic that the user selects; Inference machine carries out reverse search and reasoning according to feature coding; If the user does not select to diagnose; Then reverse search goes out with the user to select under the characteristic the compound illness characteristic of illness to upgrade the illness mark sheet, if the user selects to diagnose then to release the illness result; The illness feature extraction algorithm can calculate the illness performance degree of membership of accordingly result according to the testing result of peripheral upgrowth situation detection system, and the maximum feature extraction as illness of degree of membership is come out, and characteristic is transferred to inference machine carries out reasoning;
Data base administration calling module: comprise database storing management and KBM; The database storing management has realized the real-time storage of the detection data that peripheral growing state detection system draws and has called; The result of automatic illness reasoning is provided to user prompt module; And database management function is provided to the user, comprise interpolation, deletion, the query function of data; KBM comprises interpolation, deletion, renewal, the query function of knowledge in the knowledge base, and carries out calling of different knowledge bases of crop different bearing stage according to the growing stage signal of growing state decision-making subsystem, and this calls to the illness reasoning module and opens;
Wherein the user prompt module of different sub-systems is different, and the data base administration calling module is also different; Expert control system of the present invention links to each other with peripheral environment monitoring system, upgrowth situation detection system and environment conditioning system simultaneously, has the function of environment conditioning decision-making, growing state detection and illness diagnosis.
2. one kind is adopted the described crop greenhouse production of claim 1 expert control system to carry out crop illness method of diagnosing; It is characterized in that described diagnostic method comprises crop pest diagnosis coding knowledge representation mode and corresponding inference mechanism thereof, this method can adapt to the replacing and the illness diagnosis needs of Different Crop knowledge base;
Concrete diagnosis algorithm is following:
The illness feature extraction algorithm, the notion of the Imperfect Information Systems in this algorithm use rough set theory is summed up incomplete decision information table; Illness characteristic attribute rough set is proposed; And, collection is promptly described, as basic module to have the support collection class of identical description; Merge and constitute similar collection; Calculate the algorithm of illness characteristic blur level to derive ambiguity function, the data obfuscation that the upgrowth situation detection system is gathered forms the incomplete decision table of gathering sample and illness characteristic, and calculates that corresponding membership function value reaches that feature identification is chosen and then the purpose diagnosed of illness automatically.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210141115.3A CN102681438B (en) | 2012-05-09 | 2012-05-09 | Crop greenhouse cultivation expert control system and crop disease diagnostic method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210141115.3A CN102681438B (en) | 2012-05-09 | 2012-05-09 | Crop greenhouse cultivation expert control system and crop disease diagnostic method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102681438A true CN102681438A (en) | 2012-09-19 |
| CN102681438B CN102681438B (en) | 2014-04-30 |
Family
ID=46813515
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210141115.3A Expired - Fee Related CN102681438B (en) | 2012-05-09 | 2012-05-09 | Crop greenhouse cultivation expert control system and crop disease diagnostic method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102681438B (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104663295A (en) * | 2015-02-12 | 2015-06-03 | 上海赋民农业科技有限公司 | Midair tiny farm |
| CN105843301A (en) * | 2016-04-19 | 2016-08-10 | 柳州名品科技有限公司 | Vegetable greenhouse intelligent management platform having self-learning function |
| CN108122170A (en) * | 2017-12-21 | 2018-06-05 | 中国科学院合肥物质科学研究院 | Efficient ecological agriculture forestry planting monitors in real time and intelligent decision system |
| CN109102886A (en) * | 2018-08-20 | 2018-12-28 | 重庆柚瓣家科技有限公司 | The disease of old people reasoning diagnostic system of more reasoning pattern fusions |
| CN109119160A (en) * | 2018-08-20 | 2019-01-01 | 重庆柚瓣家科技有限公司 | The expert's system for distribution of out-patient department and its method of multiple inference mode |
| CN110069668A (en) * | 2019-04-09 | 2019-07-30 | 青海省科学技术信息研究所有限公司 | One kind is based on agriculture big data knowledge base management system and its Functional Design method |
| CN112965432A (en) * | 2021-02-03 | 2021-06-15 | 沙洲职业工学院 | Intelligent control device who uses in big-arch shelter |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1559175A (en) * | 2004-02-24 | 2005-01-05 | 深圳市宝安区农业科学技术推广中心 | Remote radio information feedback and controlable environment intelligent system for agricultural crops |
| US20050072935A1 (en) * | 2001-09-27 | 2005-04-07 | Robert Lussier | Bio-imaging and information system for scanning, detecting, diagnosing and optimizing plant health |
| CN101162384A (en) * | 2006-10-12 | 2008-04-16 | 魏珉 | Artificial intelligence plant growth surroundings regulate and control expert decision-making system |
| CN101303245A (en) * | 2008-07-02 | 2008-11-12 | 湖南大学 | System and method for monitoring greenhouse fine crops growing environment based on wireless sensor network |
-
2012
- 2012-05-09 CN CN201210141115.3A patent/CN102681438B/en not_active Expired - Fee Related
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050072935A1 (en) * | 2001-09-27 | 2005-04-07 | Robert Lussier | Bio-imaging and information system for scanning, detecting, diagnosing and optimizing plant health |
| CN1559175A (en) * | 2004-02-24 | 2005-01-05 | 深圳市宝安区农业科学技术推广中心 | Remote radio information feedback and controlable environment intelligent system for agricultural crops |
| CN101162384A (en) * | 2006-10-12 | 2008-04-16 | 魏珉 | Artificial intelligence plant growth surroundings regulate and control expert decision-making system |
| CN101303245A (en) * | 2008-07-02 | 2008-11-12 | 湖南大学 | System and method for monitoring greenhouse fine crops growing environment based on wireless sensor network |
Non-Patent Citations (1)
| Title |
|---|
| 吕永田等: "温室环境远程监控与诊断管理系统设计", 《农业网络信息》, no. 09, 30 September 2008 (2008-09-30), pages 14 - 16 * |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104663295A (en) * | 2015-02-12 | 2015-06-03 | 上海赋民农业科技有限公司 | Midair tiny farm |
| CN105843301A (en) * | 2016-04-19 | 2016-08-10 | 柳州名品科技有限公司 | Vegetable greenhouse intelligent management platform having self-learning function |
| CN108122170A (en) * | 2017-12-21 | 2018-06-05 | 中国科学院合肥物质科学研究院 | Efficient ecological agriculture forestry planting monitors in real time and intelligent decision system |
| CN109102886A (en) * | 2018-08-20 | 2018-12-28 | 重庆柚瓣家科技有限公司 | The disease of old people reasoning diagnostic system of more reasoning pattern fusions |
| CN109119160A (en) * | 2018-08-20 | 2019-01-01 | 重庆柚瓣家科技有限公司 | The expert's system for distribution of out-patient department and its method of multiple inference mode |
| CN109119160B (en) * | 2018-08-20 | 2022-08-05 | 重庆柚瓣家科技有限公司 | Expert triage system with multiple reasoning modes and method thereof |
| CN109102886B (en) * | 2018-08-20 | 2022-08-05 | 重庆柚瓣家科技有限公司 | Multi-inference mode fused geriatric disease inference diagnosis system |
| CN110069668A (en) * | 2019-04-09 | 2019-07-30 | 青海省科学技术信息研究所有限公司 | One kind is based on agriculture big data knowledge base management system and its Functional Design method |
| CN110069668B (en) * | 2019-04-09 | 2021-12-14 | 青海省科学技术信息研究所有限公司 | Agricultural big data based knowledge base management system and function design method thereof |
| CN112965432A (en) * | 2021-02-03 | 2021-06-15 | 沙洲职业工学院 | Intelligent control device who uses in big-arch shelter |
| CN112965432B (en) * | 2021-02-03 | 2023-12-15 | 沙洲职业工学院 | Intelligent control device used in greenhouse |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102681438B (en) | 2014-04-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102681438B (en) | Crop greenhouse cultivation expert control system and crop disease diagnostic method | |
| US11804016B2 (en) | Augmented reality based horticultural care tracking | |
| CN106254476A (en) | Agroecological environment information management based on Internet of Things, big data and cloud computing and monitoring method and system | |
| CN107895595B (en) | User-demand-oriented planning selection scheme recommendation method for software rapid definition intelligent environment | |
| AU2019415077A1 (en) | Information processing device, and information processing system | |
| KR20210090394A (en) | management system for smart-farm machine learning | |
| CN106650212A (en) | Intelligent plant breeding method and system based on data analysis | |
| CN117391482B (en) | Greenhouse temperature intelligent early warning method and system based on big data monitoring | |
| CN116661530B (en) | Intelligent control system and method in edible fungus industrial cultivation | |
| CN115115126A (en) | Crop disaster loss and yield prediction system and method based on knowledge graph | |
| CN118095792B (en) | Intelligent agriculture management system based on big data | |
| CN115292753A (en) | Agricultural greenhouse data tracing and management method based on block chain | |
| CN114563988A (en) | Water plant intelligent PAC adding method and system based on random forest algorithm | |
| AU2021105327A4 (en) | A computer implemented and IoT based method for increasing crop production using machine learning model | |
| Liu et al. | Evaluation of cultivated land quality using attention mechanism-back propagation neural network | |
| Morain et al. | Artificial Intelligence for Water Consumption Assessment: State of the Art Review | |
| CN116182945B (en) | Controllable agricultural greenhouse environment monitoring system and method based on wireless sensor network | |
| CN116886720A (en) | Intelligent monitoring system and method for laying hen breeding environment based on Internet of things technology | |
| CN108777850A (en) | A kind of wetland evapotranspiration real-time monitoring system based on Internet of Things | |
| CN109601443A (en) | The Puffer diseases monitoring early warning of data-driven river and self-service diagnostic assistance DSS | |
| Ahmada et al. | A Cyber-Physical System for Zero-Defect Production in Aquaponics 4.0 Learning Factory | |
| CN117557402B (en) | Wheat pest remote monitoring and early warning system based on intelligent agriculture | |
| WO2024171774A1 (en) | Information processing device, information processing method, and program | |
| Singh et al. | Precision Agriculture Monitoring System | |
| Khan et al. | Data Pre-Processing and Evaluating the Performance of Several Data Mining Methods for Predicting Irrigation Water Requirement |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20140430 Termination date: 20190509 |