CN108319134A - A kind of greenhouse environment intelligent control method based on extreme learning machine network - Google Patents
A kind of greenhouse environment intelligent control method based on extreme learning machine network Download PDFInfo
- Publication number
- CN108319134A CN108319134A CN201810035493.0A CN201810035493A CN108319134A CN 108319134 A CN108319134 A CN 108319134A CN 201810035493 A CN201810035493 A CN 201810035493A CN 108319134 A CN108319134 A CN 108319134A
- Authority
- CN
- China
- Prior art keywords
- learning machine
- control
- machine network
- data
- deviation
- 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.)
- Withdrawn
Links
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
- G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
- G05B13/04—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators
- G05B13/042—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators in which a parameter or coefficient is automatically adjusted to optimise the performance
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Greenhouses (AREA)
- Feedback Control In General (AREA)
Abstract
A kind of greenhouse environment intelligent control method based on extreme learning machine network, is related to greenhouse environment intelligent control method, and the method includes following preparation process:The collection of sample data according to the empirical data for certain crop growth that agricultural experts are provided, then passes through experiment, deviation of the acquisition crops in different growth periods;The normalized of sample data, the neural network model of foundation;Establish single extreme learning machine network-control model:Extreme learning machine network is made of input layer, hidden layer and output layer;Establish multiple limits learning machine network-control model:Multiple extreme learning machine network models are combined, multiple limits learning machine network-control model is constituted;Intelligent control is carried out to greenhouse using multiple limits learning machine network-control model, obtains green house control parameter.The present invention, which realizes, accurately controls crops in greenhouse growing environment, improves the yield and quality of crops in greenhouse.
Description
Technical field
The present invention relates to one kind, more particularly to a kind of greenhouse intelligent control side based on extreme learning machine network
Method.
Background technology
China is large agricultural country, and cultivated area of the greenhouse in China is very wide.Greenhouse can be crop growth
Suitable environment is provided, crops is made to avoid the influence of harsh weather.Intelligent plantation is realized using greenhouse, can be improved
The yield and quality of crops.China's tradition greenhouse production process needs a large amount of labour personnel to participate in, by experience management.
In fact, in the growth course of crops, temperature, humidity, intensity of illumination, CO2The factors such as concentration are to influence crop growth
Key factor.In order to improve the yield and quality of crops, it is necessary to the temperature in greenhouse, humidity, CO2Concentration etc.
Greenhouse environment parameter carries out accurate real-time control, and plant is made to be grown in always within the scope of best environmental parameter.
Currently, there are many method for environment data controlling for greenhouse, but there is control accuracies mostly high not enough, control
The problems such as effect is not ideal enough.Based on the greenhouse environment intelligent control method of extreme learning machine network, by extreme learning machine nerve
In greenhouse flower, control accuracy is higher for network application, and control effect is preferable.
Invention content
The purpose of the present invention is to provide a kind of greenhouse environment intelligent control method based on extreme learning machine network, this hairs
The bright site environment parameter by acquisition crop growth, and compare with the standard environment parameter of crop growth, it establishes
Extreme learning machine network control algorithm according to plant growth;It realizes and greenhouse environment parameter is accurately controlled.
The purpose of the present invention is what is be achieved through the following technical solutions:
A kind of greenhouse environment intelligent control method based on extreme learning machine network, the method includes following preparation process:
1)The collection of sample data:According to the empirical data for certain crop growth that agricultural experts are provided, then by testing,
Crops are obtained in the practical intensity of illumination of different growth periods and deviation, the practical soil moisture and the warp of experience intensity of illumination
Test the deviation of the soil moisture, the deviation of practical soil moisture and experience soil moisture, actual air temperature and experience air themperature
Deviation, the deviation of actual air humidity and experience air humidity, practical gas concentration lwevel and experience gas concentration lwevel
Deviation;
2)The normalized of sample data:By sampled dataIt normalizes in [0,1] section and is:
Wherein,For i-th of sampled data after normalized,For the minimum value of r sampled data,
For the maximum value of r sampled data;The neural network model of foundation;
3)Establish single extreme learning machine network-control model:Extreme learning machine network is made of input layer, hidden layer and output layer;
4)Establish multiple limits learning machine network-control model:Multiple extreme learning machine network models are combined, are constituted more
Extreme learning machine network-control model;
5)Intelligent control is carried out to greenhouse using multiple limits learning machine network-control model:Obtained each monopole is limited
The weighting coefficient of learning machine network brings multiple limits learning machine network model into, when the input of multiple limits learning machine network model is farming
The different time of object growth and the intensity of illumination deviation data of corresponding each time, soil moisture deviation data, air themperature are inclined
The input sample data of difference data, air humidity deviation data, gas concentration lwevel deviation data as extreme learning machine network
When, it can calculate cascade control and regulation amount using the built-up pattern, fan control regulated quantity, light compensating lamp control and regulation amount, dissipate
Hot device control and regulation amount, sunshading board control and regulation amount, motor control and regulation amount of leaking informaton;To cascade control and regulation amount, fan control
Regulated quantity, light compensating lamp control and regulation amount, radiator control and regulation amount, sunshading board control and regulation amount, motor control and regulation amount of leaking informaton
After each numerical value carries out anti-normalization processing, green house control parameter can be obtained.
A kind of greenhouse environment intelligent control method based on extreme learning machine network, the collection of the sample data
By intensity of illumination deviation data, soil moisture deviation data, the sky of the different time of crop growth and corresponding Each point in time
Temperature degree deviation data, the input of air humidity deviation data, gas concentration lwevel deviation data as extreme learning machine network
Sample data.
A kind of greenhouse environment intelligent control method based on extreme learning machine network, the collection of the sample data
By the corresponding cascade control and regulation amount of each deviation data of different time, fan control regulated quantity, light compensating lamp control and regulation amount,
Radiator control and regulation amount, sunshading board control and regulation amount, sample of the motor control and regulation amount of leaking informaton as extreme learning machine network
Output data.
A kind of greenhouse environment intelligent control method based on extreme learning machine network, the normalizing of the sample data
Change is handled, and r sample data after normalized is divided into two groups:One group as the common first of each single neural network model
Beginning training dataset is respectively trained each single extreme learning machine network;Another group is used as verification data collection, for verifying
The neural network model of foundation.
Advantages of the present invention is with effect:
Extreme learning machine network is applied in the control of greenhouse environment parameter by the present invention, according to the rule of plant growth, realizes
To the intelligent control of greenhouse environment parameter, crops in greenhouse growing environment is accurately controlled to realize, improves temperature
The yield and quality of room crops.
Description of the drawings
Fig. 1 is single extreme learning machine network-control model topology structure chart;
Fig. 2 is multiple limits learning machine network model topology diagram.
Specific implementation mode
The following describes the present invention in detail with reference to examples.
The present invention is realized by following steps:
One, the collection of sample data
According to the empirical data for certain crop growth that agricultural experts are provided, including needed for the different growth period of crops
Intensity of illumination, the soil moisture, soil moisture, air themperature, air humidity, the gas concentration lwevel data wanted.Again by testing,
Crops are obtained in the practical intensity of illumination of different growth periods and deviation, the practical soil moisture and the warp of experience intensity of illumination
Test the deviation of the soil moisture, the deviation of practical soil moisture and experience soil moisture, actual air temperature and experience air themperature
Deviation, the deviation of actual air humidity and experience air humidity, practical gas concentration lwevel and experience gas concentration lwevel
Deviation;And according to experiment, corresponding different intensity of illumination deviation data, soil moisture deviation data, soil moisture deviation are obtained
Cascade under data, air themperature deviation data, air humidity deviation data, gas concentration lwevel deviation data controls to adjust
Amount, fan control regulated quantity, light compensating lamp control and regulation amount, radiator control and regulation amount, sunshading board control and regulation amount, leak informaton motor
Control and regulation amount.In these data obtained, by the illumination of the different time of crop growth and corresponding Each point in time
Strength variance data, soil moisture deviation data, air themperature deviation data, air humidity deviation data, gas concentration lwevel
Input sample data of the deviation data as extreme learning machine network;By the corresponding cascade control of each deviation data of different time
Regulated quantity processed, light compensating lamp control and regulation amount, radiator control and regulation amount, sunshading board control and regulation amount, is put at fan control regulated quantity
Sample output data of the wind turbine control and regulation amount as extreme learning machine network.
Two, the normalized of sample dataIt normalizes in [0,1] section and is:
Wherein,For i-th of sampled data after normalized,For the minimum value of r sampled data,
For the maximum value of r sampled datas.
R sample data after normalized is divided into two groups:One group as the common of each single neural network model
Initial training data set is respectively trained each single extreme learning machine network;Another group is used as verification data collection, for verifying
The neural network model established.
Three, single extreme learning machine network-control model is established
Extreme learning machine network is made of input layer, hidden layer and output layer.Input layer variable is, right respectively
The parameter answered is crop growth time, the deviation of practical intensity of illumination and experience intensity of illumination, the practical soil moisture and experience
The deviation of the soil moisture, the deviation of practical soil moisture and experience soil moisture, actual air temperature and experience air themperature
Deviation, the deviation of actual air humidity and experience air humidity, practical gas concentration lwevel are inclined with experience gas concentration lwevel
Difference;Hidden layer has L number of nodes;It is cascade control and regulation amount, fan that output layer, which has 6 nodes, corresponding control parameter,
Control and regulation amount, light compensating lamp control and regulation amount, radiator control and regulation amount, sunshading board control and regulation amount, the motor that leaks informaton control are adjusted
Section amount.
Its output is denoted asIndicate maximum sampling number.Indicate ith sampling
J-th of parameter value.
Algorithmic procedure is as follows:
Take training set, whereinFor j-th of input sample data,For j-th of output sample
Data, select activation primitive for, it is 100 to select maximum node in hidden layer, it is expected that it is 0.02 to learn precision.
The determination method of the number of hidden nodes is as follows:
(1)Enable initial the number of hidden nodes;Ith for j-th of parameter is adopted
Sample value.
(2)Increase a hidden node, the number of hidden nodes;
(3)Weights are assigned at random to the hidden node newly increased;
(4)It is calculated to the node newly increased and exports weights, before updateThe output weights of a hidden node:;
(5)It calculates and increases by theError after a hidden node;
(6)WhenOr 2 norms of error EWhen, then stop, otherwise, turning(2)Continue.
After the determination of hidden layer node number, calculates network and export weights
This process is divided into following several steps:
(1)The random input layer that generates is to the weights of hidden layerWith each neuron threshold value of hidden layer;
(2)Calculate hidden layer output matrix
(3)Network hidden layer is calculated to the weights of output layer
WhenWhen being nonsingular,
WhenWhen being nonsingular,
Four, multiple limits learning machine network-control model is established
The present invention combines multiple extreme learning machine network models, constitutes multiple limits learning machine network-control model.It is more
Total output of extreme learning machine network-control model is the weighted sum of each single neural network output, i.e.,:
The wherein input data matrix of X multiple neural networks,For multiple neural network output vector,For multiple neural network
Prediction model,For single extreme learning machine network number,It isA list extreme learning machine Network Prediction Model,For
TheThe non-negative weight vector of a list neural network,, meet regression nature, i.e.,:。
The present invention is for weight distributions such as the determination uses of each weight coefficient, i.e.,。
Five, intelligent control is carried out to greenhouse using multiple limits learning machine network-control model
It brings the weighting coefficient of obtained each single extreme learning machine network into multiple limits learning machine network model, works as multiple limits
The input of learning machine network model is intensity of illumination deviation data, the soil of the different time and corresponding each time of crop growth
Temperature deviation data, air themperature deviation data, air humidity deviation data, gas concentration lwevel deviation data are as the limit
When the input sample data of habit machine network, cascade control and regulation amount can be calculated using the built-up pattern, fan control is adjusted
Amount, light compensating lamp control and regulation amount, radiator control and regulation amount, sunshading board control and regulation amount, motor control and regulation amount of leaking informaton;To water
Curtain control and regulation amount, fan control regulated quantity, light compensating lamp control and regulation amount, radiator control and regulation amount, sunshading board control to adjust
It measures, after each numerical value progress anti-normalization processing of motor control and regulation amount of leaking informaton, green house control parameter can be obtained.
Embodiment:
The collection of sample data according to the empirical data for certain crop growth that agricultural experts are provided, and according to experiment, obtains
To intensity of illumination deviation data, soil moisture deviation data, the sky of the different time and corresponding Each point in time of crop growth
Temperature degree deviation data, the input of air humidity deviation data, gas concentration lwevel deviation data as extreme learning machine network
Sample data;By the corresponding cascade control and regulation amount of each deviation data of different time, fan control regulated quantity, light compensating lamp control
Regulated quantity processed, radiator control and regulation amount, sunshading board control and regulation amount, motor control and regulation amount of leaking informaton are as extreme learning machine net
The sample output data of network.
The normalized of sample data, by sampled dataIt normalizes in [0,1] section and is:
Wherein,It isA sampled data after normalized,For the minimum value of 200 sampled datas,For the maximum value of 200 sampled datas.
200 sample datas after normalized are divided into two groups:One group of 150 sample data is limited as each monopole
The common initial training data set of learning machine network model is respectively trained each single extreme learning machine network;Another group 50
A sample data, as verification data collection, for verifying established neural network model.
Single extreme learning machine network-control model is established, the corresponding parameter of extreme learning machine network input layer is crop
Growth time, the deviation of practical intensity of illumination and experience intensity of illumination, the deviation of the practical soil moisture and the experience soil moisture,
The deviation of practical soil moisture and experience soil moisture, actual air temperature and the deviation of experience air themperature, actual air are wet
Deviation, the deviation of practical gas concentration lwevel and experience gas concentration lwevel of degree and experience air humidity;Output layer has 6 sections
Point, corresponding control parameter are cascade control and regulation amount, fan control regulated quantity, light compensating lamp control and regulation amount, radiator
Control and regulation amount, sunshading board control and regulation amount, motor control and regulation amount of leaking informaton.
Algorithmic procedure is as follows:
Take training set, whereinFor j-th of input sample data,It is j-th
Export sample data, select activation primitive for, it is 100 to select maximum node in hidden layer, it is expected that learning
It is 0.02 to practise precision.
The determination method of the number of hidden nodes is as follows:
(1)Enable initial the number of hidden nodes, error;
(2)Increase a hidden node, the number of hidden nodes;
(3)Weights are assigned at random to the hidden node newly increased;
(4)It is calculated to the node newly increased and exports weights;
Wherein,
(5)Before updateThe output weights of a hidden node:;
(6)It calculates and increases by theError after a hidden node;
(7)WhenOr 2 norms of error EWhen, then stop, otherwise, turning(2)Continue.
After the determination of hidden layer node number, network exports weightsCalculating process it is as follows:
This process is divided into following several steps:
(1)The random input layer that generates is to the weights of hidden layerWith each neuron threshold value of hidden layer;
(2)Calculate hidden layer output matrix
(3)Network hidden layer is calculated to the weights of output layer
WhenWhen being nonsingular,
WhenWhen being nonsingular,
,
Multiple limits learning machine network-control model is established, the present embodiment combines 3 single extreme learning machine network models,
Constitute multiple limits learning machine network-control model.Total output of multiple limits learning machine network-control model is that each single neural network is defeated
The weighted sum gone out, i.e.,:
In the present embodiment, weight coefficient。
Intelligent control is carried out to greenhouse using multiple limits learning machine network-control model, by obtained each monopole
The weighting coefficient of limit learning machine network brings multiple limits learning machine network model into, when the input of multiple limits learning machine network model is agriculture
The different time of plant growth and the intensity of illumination deviation data of corresponding each time, soil moisture deviation data, air themperature
The input sample number of deviation data, air humidity deviation data, gas concentration lwevel deviation data as extreme learning machine network
According to when, using the built-up pattern can calculate cascade control and regulation amount, fan control regulated quantity, light compensating lamp control and regulation amount,
Radiator control and regulation amount, sunshading board control and regulation amount, motor control and regulation amount of leaking informaton;To cascade control and regulation amount, fan control
Regulated quantity processed, light compensating lamp control and regulation amount, radiator control and regulation amount, sunshading board control and regulation amount, the motor that leaks informaton control to adjust
After measuring each numerical value progress anti-normalization processing, green house control parameter can be obtained.
Claims (4)
1. a kind of greenhouse environment intelligent control method based on extreme learning machine network, which is characterized in that the method includes with
Lower preparation process:
1)The collection of sample data:According to the empirical data for certain crop growth that agricultural experts are provided, then by testing,
Crops are obtained in the practical intensity of illumination of different growth periods and deviation, the practical soil moisture and the warp of experience intensity of illumination
Test the deviation of the soil moisture, the deviation of practical soil moisture and experience soil moisture, actual air temperature and experience air themperature
Deviation, the deviation of actual air humidity and experience air humidity, practical gas concentration lwevel and experience gas concentration lwevel
Deviation;
2)The normalized of sample data:By sampled dataIt normalizes in [0,1] section and is:
Wherein,It isA sampled data after normalized,For the minimum value of r sampled data,
For the maximum value of r sampled data;The neural network model of foundation;
3)Establish single extreme learning machine network-control model:Extreme learning machine network is made of input layer, hidden layer and output layer;
4)Establish multiple limits learning machine network-control model:Multiple extreme learning machine network models are combined, are constituted more
Extreme learning machine network-control model;
5)Intelligent control is carried out to greenhouse using multiple limits learning machine network-control model:Obtained each monopole is limited
The weighting coefficient of learning machine network brings multiple limits learning machine network model into, when the input of multiple limits learning machine network model is farming
The different time of object growth and the intensity of illumination deviation data of corresponding each time, soil moisture deviation data, air themperature are inclined
The input sample data of difference data, air humidity deviation data, gas concentration lwevel deviation data as extreme learning machine network
When, it can calculate cascade control and regulation amount using the built-up pattern, fan control regulated quantity, light compensating lamp control and regulation amount, dissipate
Hot device control and regulation amount, sunshading board control and regulation amount, motor control and regulation amount of leaking informaton;To cascade control and regulation amount, fan control
Regulated quantity, light compensating lamp control and regulation amount, radiator control and regulation amount, sunshading board control and regulation amount, motor control and regulation amount of leaking informaton
After each numerical value carries out anti-normalization processing, green house control parameter can be obtained.
2. a kind of greenhouse environment intelligent control method based on extreme learning machine network according to claim 1, feature
It is, the collection of the sample data is by the intensity of illumination variation of the different time of crop growth and corresponding Each point in time
According to, soil moisture deviation data, air themperature deviation data, air humidity deviation data, gas concentration lwevel deviation data make
For the input sample data of extreme learning machine network.
3. a kind of greenhouse environment intelligent control method based on extreme learning machine network according to claim 2, feature
It is, the collection of the sample data is by the corresponding cascade control and regulation amount of each deviation data of different time, fan control
Regulated quantity, light compensating lamp control and regulation amount, radiator control and regulation amount, sunshading board control and regulation amount, motor control and regulation amount of leaking informaton
Sample output data as extreme learning machine network.
4. a kind of greenhouse environment intelligent control method based on extreme learning machine network according to claim 1, feature
It is, the normalized of the sample data, r sample data after normalized is divided into two groups:One group as each
The common initial training data set of single neural network model is respectively trained each single extreme learning machine network;Another group
As verification data collection, for verifying established neural network model.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810035493.0A CN108319134A (en) | 2018-01-15 | 2018-01-15 | A kind of greenhouse environment intelligent control method based on extreme learning machine network |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810035493.0A CN108319134A (en) | 2018-01-15 | 2018-01-15 | A kind of greenhouse environment intelligent control method based on extreme learning machine network |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108319134A true CN108319134A (en) | 2018-07-24 |
Family
ID=62893783
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810035493.0A Withdrawn CN108319134A (en) | 2018-01-15 | 2018-01-15 | A kind of greenhouse environment intelligent control method based on extreme learning machine network |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108319134A (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109856973A (en) * | 2019-02-25 | 2019-06-07 | 山东省农业机械科学研究院 | A kind of Technique for Controlling Greenhouse Environment and system based on fuzzy neural network |
WO2020056812A1 (en) * | 2018-09-21 | 2020-03-26 | 苏州数言信息技术有限公司 | Environmental parameter weight determining method and system for evaluating indoor environmental quality |
CN111346688A (en) * | 2018-12-24 | 2020-06-30 | 航天信息股份有限公司 | Wheat dampening control method and device |
CN111524023A (en) * | 2020-04-07 | 2020-08-11 | 中国农业大学 | Greenhouse adjusting method and system |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101315544A (en) * | 2007-06-01 | 2008-12-03 | 上海电机学院 | Greenhouse intelligent control method |
KR101173823B1 (en) * | 2011-07-01 | 2012-08-20 | 연세대학교 산학협력단 | System and method for predicting an energy consumption of multi-family housing |
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 |
CN105740619A (en) * | 2016-01-28 | 2016-07-06 | 华南理工大学 | On-line fault diagnosis method of weighted extreme learning machine sewage treatment on the basis of kernel function |
CN107168066A (en) * | 2017-06-23 | 2017-09-15 | 太原理工大学 | A kind of greenhouse self-adaptation control method |
-
2018
- 2018-01-15 CN CN201810035493.0A patent/CN108319134A/en not_active Withdrawn
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101315544A (en) * | 2007-06-01 | 2008-12-03 | 上海电机学院 | Greenhouse intelligent control method |
KR101173823B1 (en) * | 2011-07-01 | 2012-08-20 | 연세대학교 산학협력단 | System and method for predicting an energy consumption of multi-family housing |
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 |
CN105740619A (en) * | 2016-01-28 | 2016-07-06 | 华南理工大学 | On-line fault diagnosis method of weighted extreme learning machine sewage treatment on the basis of kernel function |
CN107168066A (en) * | 2017-06-23 | 2017-09-15 | 太原理工大学 | A kind of greenhouse self-adaptation control method |
Non-Patent Citations (4)
Title |
---|
QI LIU 等: "A WSN一Based Prediction Model of Microclimate in a Greenhouse Using Extreme Learning Approaches", 《ICACT TRANSACTIONS ON ADVANCED COMMUNICATIONS TECHNOLOGY (TACT)》 * |
张园园: "温室小气候建模和控制策略的研究", 《中国优秀硕士学位论文全文数据库农业科技辑》 * |
江沸菠 等: "《基于神经网络的混合非线性电阻率反演成像》", 31 October 2015, 中南大学出版社 * |
邹伟东 等: "基于改进型极限学习机的日光温室温湿度预测与验证", 《农业工程学报》 * |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2020056812A1 (en) * | 2018-09-21 | 2020-03-26 | 苏州数言信息技术有限公司 | Environmental parameter weight determining method and system for evaluating indoor environmental quality |
CN111346688A (en) * | 2018-12-24 | 2020-06-30 | 航天信息股份有限公司 | Wheat dampening control method and device |
CN111346688B (en) * | 2018-12-24 | 2021-08-24 | 航天信息股份有限公司 | Wheat dampening control method and device |
CN109856973A (en) * | 2019-02-25 | 2019-06-07 | 山东省农业机械科学研究院 | A kind of Technique for Controlling Greenhouse Environment and system based on fuzzy neural network |
CN111524023A (en) * | 2020-04-07 | 2020-08-11 | 中国农业大学 | Greenhouse adjusting method and system |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108983849A (en) | It is a kind of to utilize compound extreme learning machine ANN Control greenhouse method | |
CN108319134A (en) | A kind of greenhouse environment intelligent control method based on extreme learning machine network | |
CN104155925B (en) | Henhouse micro climatic environment intelligent control system based on wireless sensor network | |
CN107909149A (en) | A kind of Temperature in Greenhouse Forecasting Methodology based on Genetic BP Neutral Network | |
CN107622361B (en) | A kind of mathematical modeling and its control method for edible mushroom greenhouse-environment | |
CN109634098A (en) | A kind of fattening house environment conditioning system and method | |
WO2023179167A1 (en) | Crop irrigation water demand prediction method based on aquacrop model and svr | |
CN110119169B (en) | Tomato greenhouse temperature intelligent early warning system based on minimum vector machine | |
CN107329511A (en) | Based on the vegetable aquaculture luminous environment high efficiency regulatory method and system between suitable root warm area | |
CN110244559A (en) | A kind of greenhouse intelligent regulation method based on agriculture solar term empirical data | |
CN107219759B (en) | Greenhouse environment control method and device | |
CN111915062B (en) | Greenhouse crop water demand regulation and control method with water utilization rate and photosynthesis rate being coordinated | |
CN107037728A (en) | Greenhouse optimal control method based on multiple objective gray particle cluster algorithm | |
CN108919768A (en) | A kind of agricultural management system and method | |
Ting et al. | Management of CO2 in a tomato greenhouse using WSN and BPNN techniques | |
CN107145941A (en) | The real-time dynamic acquisition method of light requirement based on optimal light quality and photon flux density | |
Jiang et al. | WSN-based control system of CO2 concentration in greenhouse | |
CN109213240A (en) | A kind of strawberry greenhouse wireless monitor and control system based on self adaptive control | |
CN104346913A (en) | Remote tender branch cutting control system | |
Wang et al. | Real-time monitoring of environmental information and modeling of the photosynthetic rate of tomato plants under greenhouse conditions | |
CN113625806B (en) | Optimal-efficiency greenhouse plant regulation and control method for carbon neutralization demand | |
CN112783228B (en) | Neural network-based large-scale greenhouse PD control system and method | |
Zhang et al. | Modeling and simulation of temperature control system in plant factory using energy balance. | |
CN112099557A (en) | Internet-based household plant planting method and system | |
CN115453868B (en) | Full-growth-period light intensity regulation and control method based on tomato light response difference characteristics |
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 | ||
WW01 | Invention patent application withdrawn after publication | ||
WW01 | Invention patent application withdrawn after publication |
Application publication date: 20180724 |