CN103903068A - Greenhouse energy forecasting method based on hybrid optimization algorithm - Google Patents

Greenhouse energy forecasting method based on hybrid optimization algorithm Download PDF

Info

Publication number
CN103903068A
CN103903068A CN201410144270.XA CN201410144270A CN103903068A CN 103903068 A CN103903068 A CN 103903068A CN 201410144270 A CN201410144270 A CN 201410144270A CN 103903068 A CN103903068 A CN 103903068A
Authority
CN
China
Prior art keywords
particle
greenhouse
population
value
formula
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
Application number
CN201410144270.XA
Other languages
Chinese (zh)
Other versions
CN103903068B (en
Inventor
陈教料
陈教选
胥芳
艾青林
赵江武
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Zhejiang University of Technology ZJUT
Original Assignee
Zhejiang University of Technology ZJUT
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Zhejiang University of Technology ZJUT filed Critical Zhejiang University of Technology ZJUT
Priority to CN201410144270.XA priority Critical patent/CN103903068B/en
Publication of CN103903068A publication Critical patent/CN103903068A/en
Application granted granted Critical
Publication of CN103903068B publication Critical patent/CN103903068B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Abstract

The invention discloses a greenhouse energy forecasting method based on a hybrid optimization algorithm. The greenhouse forecasting method based on the hybrid optimization algorithm comprises the following steps that (1), a differential equation of temperature inside a greenhouse is set; (2), parameters are initialized; (3), a population is initialized, and the initial values of the parameters needing recognizing are generated randomly; (4), gen is made to be 1; (5), if gen is smaller than or equal to gens_max, the step (6) is carried out, and if gen is greater than gens_max, the step (15) is carried out; (6), k is made to be 1; (7), if k is smaller than or equal to max_k, the step (8) is carried out, or the step (10) is carried out; (8), a current optimal solution and a globally optimal solution are obtained; (9), k is made to be k+1, and the step (7) is carried out again; (10), pop_size grains are selected by utilization of a preferred function; (11), information of reserved M grains is used for regenerating a population of the GA; (12), the grains obtained in the step (11) are used for intersection and variation of the GA; (13), the pop_size-M grains obtained by the GA and the reserved M grains of the PSO are combined to be pop_size new populations; (14), gen is made to be gen+1, and then the step (5) is carried out; (15), the minimum fitness function value and the parameters are output finally, and the forecast energy value of the greenhouse is output.

Description

A kind of greenhouse energy predicting method of hybrid optimization algorithm
Technical field
The invention belongs to agricultural greenhouse Environment Design and control technology field, be applicable to be difficult to set up the greenhouse energy predicting field of mathematical models.Specifically, relate to a kind of based on adaptive particle swarm optimization algorithm ( particle Swarm Optimization, hereinafter to be referred as pSO) and genetic algorithm ( genetic Algorithm, hereinafter to be referred as gA) the greenhouse energy predicting method of hybrid optimization algorithm.
Background technology
Industrialized agriculture is one of key industrial sector of modern agriculture, is the mark of a national agricultural development level, is the inevitable requirement of an agriculture featuring high yields, fine quality and high efficiency.The important content that hothouse production industry is produced as industrialized agriculture, causes people's common concern gradually.
Greenhouse is the complication system with non-linear, randomness, strong coupling and the feature such as uncertain.The object of greenhouse modeling is mainly the needs that meet the aspects such as Greenhouse System emulation, design, prediction, control (optimization control and adaptive control) and decision-making.The main method of greenhouse modeling has two kinds of modelling by mechanism based on physical process and methods of testing.Modelling by mechanism based on physical process is to utilize unsteady-state heat transfer mass transfer to obtain describing the differential equation of greenhouse dynamic process, obtains environment dynamic model by solving the differential equation in the solution under certain boundary conditions.But in the time that physical parameter is more, some physical parameter wherein will be difficult to measure and some parameter can change according to the growth of houseplant.Therefore be difficult to set up energy mathematical model more accurately by the modelling by mechanism method of physical process completely.Method of testing refers to the mathematical model of utilizing information that inputoutput data provides to carry out process of establishing, also referred to as System Discrimination modeling.For very complicated system, it is more difficult setting up the mechanism model that each parameter has physical significance.So from experimental data, set up the model of input/output relation that can reflect system, although do not there is clear and definite physical significance, can be the Exact Design of system with control conduct with reference to foundation.
The present invention is the modeling method combining with System Discrimination modeling method according to modelling by mechanism, adopts pSO- gAhybrid optimization algorithm carries out identification acquisition to being difficult to definite parameter in physical model.Greenhouse environment control system need to be according to the variation of greenhouse climate environment, controls topworks and regulates accordingly.When indoor air temperature is too low, need to utilize heating system heating; When indoor air temperature is too high, need to control the work such as ventilating opening, shading system, vent fan or evaporation-cooled device, avoid overheated.Therefore, energy predicting method in greenhouse can be greenhouse design and controls provides theoretical reference.
Summary of the invention
In order to overcome existing greenhouse energy predicting method precision deficiency and consuming time of a specified duration, the invention provides a kind of greenhouse energy predicting method of the hybrid optimization algorithm based on self-adaptation PSO and GA.
The technical solution adopted for the present invention to solve the technical problems is:
A greenhouse energy predicting method for hybrid optimization algorithm, is characterized in that comprising the following steps:
Step 1: the differential equation of setting up warm indoor temperature:
(I)
In formula: for the density of dry air, for the volume in greenhouse, for airborne thermal content, , be respectively the indoor and outdoor temperature in greenhouse, for the surface area of chamber covering material, for greenhouse surface area, for outdoor radiosity, for the total light transmittance of cladding material, the energy providing for greenhouse heat source of heat-supply system, for the emissivity between glass surface and air, for stefan-Boltzmanconstant, for sky temperature, for the heat transfer coefficient of cladding material, for thermal screen insulation correction factor, for plant canopy leaf area index, for the temperature of canopy leaf surface, for boundary layer aerodynamics impedance, being time series, is the adjacent seasonal effect in time series time interval;
Wherein, will , with as the object of parameter adjustment identification;
Step 2: the initialization of parameter, comprises total number of particles pop_size, preferentially function threshold values , gAin initial crossing-over rate and aberration rate , the scale-up factor of crossing-over rate and aberration rate cwith m, maximum particle rapidity v_max, pSO_GAhybrid optimization algorithm is embedded pSOthe population algebraically of algorithm max_k, pSO_GAthe population algebraically of hybrid optimization algorithm gens_max;
Step 3: initialization population, generates even number particle composition population by population scale and constraint condition random p (t); Meanwhile, the random parameter that needs identification that generates , with initial value;
Step 4: order gen=1, wherein genthe algebraically of genetic manipulation;
Step 5: if , perform step 6, otherwise exit to step 15;
Step 6: order k=1, wherein krepresent pSO_GAhybrid optimization algorithm is embedded pSOthe current population algebraically of algorithm;
Step 7: if , perform step 8, otherwise execution step 10;
Step 8: algorithm is generated , , use with interior extraneous greenhouse variable input mATLABin software sIMULINKthe Greenhouse model the inside that functional module is set up, obtains out current portion optimum solution and globally optimal solution; Then upgrade speed and the positional information of population according to formula II, III;
(II)
(III)
In formula: iindividual particle is kspeed in inferior Evolution of Population, iindividual particle is klocally optimal solution after inferior evolution, globally optimal solution, iindividual particle is kposition in inferior evolution, wconstraint, with respectively the study factor and the social factor, with it is the random number between [0,1];
Order = , then obtain their value according to formula IV:
(IV)
In formula: c is constant, and c>2;
Step 9: order , return to step 7;
Step 10: will pop_sizeindividual particle carries out arrangement from small to large by the value of fitness function, then utilizes the preferentially function of formula V to select: if the value of the fitness function of some particles j (i)be more than or equal to selected threshold values, this particle is bad class particle, gives up; If the value of the fitness function of some particles j (i)be less than selected threshold values, this particle is excellent class particle, retains; Finally can obtain mthe particle of individual fitness function value in selected threshold values;
(V)
In formula: represent the ithe preferentially function of individual particle, j (i)represent the ithe value of the fitness function of individual particle, represent selected threshold values;
Step 11: by what retain mthe information of individual particle is brought formula VI into and is regenerated gApopulation;
(VI)
In formula: ithe selecteed possibility of individual particle, ithe value of the fitness function of individual particle, mthe size of population, be all particles in population fitness function value and; Wherein fitness function can represent with formula VII;
(VII)
tindividual time series ithe output energy of individual calculating particles, tthe output energy of individual time series experiment actual measurement, nfor seasonal effect in time series maximal value;
Step 12: will with bring formula VIII and IX into, obtain crossing-over rate and aberration rate now:
(VIII)
(IX)
In formula: genthe algebraically of genetic manipulation, gens_maxbe pSO_GAthe population algebraically of hybrid optimization algorithm maximum, with respectively initial crossing-over rate and aberration rate, cwith mbe respectively the scale-up factor of crossing-over rate and aberration rate, with respectively genthe crossing-over rate in generation and aberration rate;
Then the particle, step 11 being obtained for gAcrossover and mutation;
Step 13: will gAobtain pop_size-Mindividual particle and pSOretain mindividual particle is combined into number pop_sizenew population;
Step 14: order , execution step 5;
Step 15: finally export the parameter that needs identification in minimum fitness function value and model , with ,and the energy value of greenhouse prediction.
Superiority of the present invention is mainly manifested in:
1. self-adaptation pSO-GAhybrid optimization algorithm has fast convergence rate, more easily tends to the feature of globally optimal solution;
2. based on self-adaptation pSO-GAthe greenhouse energy predicting method of hybrid optimization algorithm is more accurate and still less consuming time aspect greenhouse energy predicting.
brief description of the drawings
Be further described in detail below in conjunction with accompanying drawing and embodiments of the present invention.
Fig. 1 is greenhouse energy predicting procedure figure.
Fig. 2 is based on self-adaptation pSO-GAthe greenhouse energy predicting method flow diagram of hybrid optimization algorithm.
Fig. 3 is greenhouse energy predicting graph curve.
Embodiment
With reference to Fig. 1, in the present embodiment, greenhouse is to connect a glasshouse, employing nanometer antimony-doped stannic oxide ( aTO) glass of film, to reduce the heat interchange inside and outside greenhouse, play good energy-saving effect.It is high that unlatching at night in winter has reflectivity, and the aluminium foil thermal screen that emissivity is low reduces thermal loss.Inside greenhouse adopts fan coil and low-temperature hot water floor radiation heating system as the end heating plant in greenhouse, adopts the medium of circulating hot water as heat transport.The present embodiment carries out energy predicting by following steps:
Step 1: according to radiant heat exchange, heat conduction energy exchange, mass-and heat-transfer energy exchange and crop latent heat and Exchange of apparent heat etc., set up the differential equation of greenhouse indoor temperature.And according to the differential equation, adopt high-order Runge Kutta method for solving to change into the energy equation of greenhouse.The energy equation of greenhouse can represent that independent variable is solar radiation solar, outdoor temperature and humidity relevant, and the heat transfer coefficient of greenhouse geometrical structure parameter, cladding material , thermal screen insulation correction factor , be plant canopy leaf area index relevant expression formulas such as (leaf area/surface areas).Greenhouse self-energy predictive equation can be expressed as: (I)
Wherein, be 1.2 ; be 64512 ; be 1008 ; Inside and outside, greenhouse temperature , collected by temperature sensor; for the surface area of chamber covering material, value is 11470 ; for greenhouse surface area, value is 8064 ; for outdoor radiosity, can be obtained by weather station sensor collection; for the total light transmittance of cladding material, value is 0.75; the energy (the energy difference that different greenhouse heating systems provides) providing for greenhouse heat source of heat-supply system, for the emissivity between glass surface and air, numerical value is 0.82, for stefan-Boltzmanconstant , for sky temperature, can pass through 183+0.34 ( + 273) calculate and obtain; for the heat transfer coefficient of cladding material; for thermal screen insulation correction factor; for plant canopy leaf area index; for the temperature of canopy leaf surface, the indoor temperature before available 1 hour substitutes; for boundary layer aerodynamics impedance, value is 290 ; tbe time series, each seasonal effect in time series interval time is 5 minutes;
Wherein, greenhouse geometrical structure parameter can be determined and can calculate according to greenhouse structure, outdoor solar radiation solar,outdoor temperature and outside humidity can be obtained by the collection of outdoor weather station, and , with difficult parameters, to measure, and can change along with the situation such as dust and plant growth on cladding material, therefore , with as the object of parameter adjustment identification;
Step 2: the initialization of parameter, comprises total number of particles pop_size=20, preferentially function threshold values =250000, gAin initial crossover and mutation probability =0.6 He =0.01, the scale-up factor of crossing-over rate and aberration rate c=0.5with m=0.005, maximum particle rapidity v_max=7, pSO_GAhybrid optimization algorithm is embedded pSOthe population algebraically of algorithm max_k=10, pSO_GAthe population algebraically of hybrid optimization algorithm gens_max=10;
Step 3: initialization population, generates even number particle composition population by population scale and constraint condition random p (t).Meanwhile, the random parameter that needs identification that generates , with initial value, wherein for the random number between [3,6], for the random number between [0.5,1], for the random number between [4,8];
Step 4: order gen=1, wherein genthe algebraically of genetic manipulation;
Step 5: if , perform step 6, otherwise exit to step 15;
Step 6: order k=1, wherein krepresent pSO_GAhybrid optimization algorithm is embedded pSOthe current population algebraically of algorithm;
Step 7: if , perform step 8, otherwise execution step 10;
Step 8: algorithm is generated , , use with interior extraneous greenhouse variable input mATLABin software sIMULINKthe Greenhouse model the inside that functional module is set up, draws locally optimal solution and globally optimal solution; Then upgrade speed and the positional information of population according to formula II, III;
(II)
(III)
In formula: iindividual particle is kspeed in inferior Evolution of Population, iindividual particle is klocally optimal solution after inferior evolution, globally optimal solution, iindividual particle is kposition in inferior evolution, wbe constraint, it can ensure pSOsearch in algorithm can finally be tending towards certain a bit, with respectively the study factor and the social factor, with it is the random number between [0,1];
Order = , then obtain their value according to formula IV:
(IV)
In formula: c is constant, and c>2;
Step 9: order , return to step 7;
Step 10: will pop_sizeindividual particle carries out arrangement from small to large by the value of fitness function, then utilizes the preferentially function of formula V to select: if the value of the fitness function of some particles j (i)be more than or equal to selected threshold values, this particle is bad class particle, gives up; If the value of the fitness function of some particles j (i)be less than selected threshold values, this particle is excellent class particle, retains; Finally can obtain mthe particle of individual fitness function value in selected threshold values;
(V)
In formula: represent the ithe preferentially function of individual particle, j (i)represent the ithe value of the fitness function of individual particle, represent selected threshold values;
Step 11: by what retain mthe information of individual particle is brought formula VI into and is regenerated gApopulation; Here we define mthe selecteed possibility of particle that in individual particle, the value of fitness function is large is large, and the reason of doing is like this that this class particle is " the bad class particle " of the excellent class particle the inside of definition in step 10, and we wish that these particles are more chosen and carry out gAoperation, prevents that algorithm is absorbed in local optimum convergence;
(VI)
In formula: ithe selecteed possibility of individual particle, ithe value of the fitness function of individual particle, mthe size of population, be all particles in population fitness function value and; Wherein fitness function can represent with formula VII;
(VII)
tindividual time series ithe output energy of individual calculating particles, tthe output energy of individual time series experiment actual measurement, nfor seasonal effect in time series maximal value;
Step 12: will with bring formula VIII and IX into, obtain crossing-over rate and aberration rate now:
(VIII)
(IX)
In formula: genthe algebraically of genetic manipulation, gens_maxbe pSO_GAthe population algebraically of hybrid optimization algorithm maximum, with respectively initial crossing-over rate and aberration rate, cwith mbe respectively the scale-up factor of crossing-over rate and aberration rate, with respectively genthe crossing-over rate in generation and aberration rate;
Then the particle that, we obtain step 11 for gAcrossover and mutation;
Step 13: will gAobtain pop_size-Mindividual particle and pSOretain mindividual particle is combined into number pop_sizenew population;
Step 14: order , execution step 5;
Step 15: finally export minimum fitness function value and be 180000 and model in need the parameter of identification , with respectively 4.5,0.6,5.8, wherein the heat transfer coefficient of cladding material, thermal screen insulation correction factor, plant canopy leaf area index, i.e. leaf area/surface area.
The present embodiment conducts in-depth analysis in the combination thought of particle cluster algorithm and genetic algorithm and combination, and around population self-adaptation segmentation strategy, crossover operator is chosen and the probability parameter selection that makes a variation etc. is optimized algorithm design.By introduce the crossover operator of genetic algorithm in particle cluster algorithm, make paired particle can intercourse information, so that particle has had to the ability of new search volume flight.The mutation operator of introducing genetic algorithm in particle cluster algorithm strengthens particle cluster algorithm and jumps out local ability.By case verification new gA-PSOthe global convergence of hybrid optimization algorithm.Sample result shows self-adaptation gA-PSOhybrid optimization algorithm has possessed good ability of searching optimum, has also had stronger local search ability simultaneously.Can obtain faster greenhouse energy predicting value, and contrast actual value, value and the actual value of algorithm identification are more or less the same.
What more than enumerate is only specific embodiments of the invention.Obviously, the invention is not restricted to above embodiment, can also have many distortion.All distortion that those of ordinary skill in the art can also directly derive or associate from content disclosed by the invention, all should think protection scope of the present invention.

Claims (1)

1. a greenhouse energy predicting method for hybrid optimization algorithm, is characterized in that comprising the following steps:
Step 1: the differential equation of setting up warm indoor temperature:
(I)
In formula: for the density of dry air, for the volume in greenhouse, for airborne thermal content, , be respectively the indoor and outdoor temperature in greenhouse, for the surface area of chamber covering material, for greenhouse surface area, for outdoor radiosity, for the total light transmittance of cladding material, the energy providing for greenhouse heat source of heat-supply system, for the emissivity between glass surface and air, for stefan-Boltzmanconstant, for sky temperature, for the heat transfer coefficient of cladding material, for thermal screen insulation correction factor, for plant canopy leaf area index, for the temperature of canopy leaf surface, for boundary layer aerodynamics impedance, being time series, is the adjacent seasonal effect in time series time interval;
Wherein, will , with as the object of parameter adjustment identification;
Step 2: the initialization of parameter, comprises total number of particles pop_size, preferentially function threshold values , gAin initial crossing-over rate and aberration rate , the scale-up factor of crossing-over rate and aberration rate cwith m, maximum particle rapidity v_max, pSO_GAhybrid optimization algorithm is embedded pSOthe population algebraically of algorithm max_k, pSO_GAthe population algebraically of hybrid optimization algorithm gens_max;
Step 3: initialization population, generates even number particle composition population by population scale and constraint condition random p (t); Meanwhile, the random parameter that needs identification that generates , with initial value;
Step 4: order gen=1, wherein genthe algebraically of genetic manipulation;
Step 5: if , perform step 6, otherwise exit to step 15;
Step 6: order k=1, wherein krepresent pSO_GAhybrid optimization algorithm is embedded pSOthe current population algebraically of algorithm;
Step 7: if , perform step 8, otherwise execution step 10;
Step 8: algorithm is generated , , use with interior extraneous greenhouse variable input mATLABin software sIMULINKthe Greenhouse model the inside that functional module is set up, obtains out current portion optimum solution and globally optimal solution; Then upgrade speed and the positional information of population according to formula II, III;
(II)
(III)
In formula: iindividual particle is kspeed in inferior Evolution of Population, iindividual particle is klocally optimal solution after inferior evolution, globally optimal solution, iindividual particle is kposition in inferior evolution, wconstraint, with respectively the study factor and the social factor, with it is the random number between [0,1];
Order = , then obtain their value according to formula IV:
(IV)
In formula: c is constant, and c>2;
Step 9: order , return to step 7;
Step 10: will pop_sizeindividual particle carries out arrangement from small to large by the value of fitness function, then utilizes the preferentially function of formula V to select: if the value of the fitness function of some particles j (i)be more than or equal to selected threshold values, this particle is bad class particle, gives up; If the value of the fitness function of some particles j (i)be less than selected threshold values, this particle is excellent class particle, retains; Finally can obtain mthe particle of individual fitness function value in selected threshold values;
(V)
In formula: represent the ithe preferentially function of individual particle, j (i)represent the ithe value of the fitness function of individual particle, represent selected threshold values;
Step 11: by what retain mthe information of individual particle is brought formula VI into and is regenerated gApopulation;
(VI)
In formula: ithe selecteed possibility of individual particle, ithe value of the fitness function of individual particle, mthe size of population, be all particles in population fitness function value and; Wherein fitness function can represent with formula VII;
(VII)
tindividual time series ithe output energy of individual calculating particles, tthe output energy of individual time series experiment actual measurement, nfor seasonal effect in time series maximal value;
Step 12: will with bring formula VIII and IX into, obtain crossing-over rate and aberration rate now:
(VIII)
(IX)
In formula: genthe algebraically of genetic manipulation, gens_maxbe pSO_GAthe population algebraically of hybrid optimization algorithm maximum, with respectively initial crossing-over rate and aberration rate, cwith mbe respectively the scale-up factor of crossing-over rate and aberration rate, with respectively genthe crossing-over rate in generation and aberration rate;
Then the particle, step 11 being obtained for gAcrossover and mutation;
Step 13: will gAobtain pop_size-Mindividual particle and pSOretain mindividual particle is combined into number pop_sizenew population;
Step 14: order , execution step 5;
Step 15: finally export the parameter that needs identification in minimum fitness function value and model , with ,and the energy value of greenhouse prediction.
CN201410144270.XA 2014-04-11 2014-04-11 Greenhouse energy forecasting method based on hybrid optimization algorithm Active CN103903068B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201410144270.XA CN103903068B (en) 2014-04-11 2014-04-11 Greenhouse energy forecasting method based on hybrid optimization algorithm

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201410144270.XA CN103903068B (en) 2014-04-11 2014-04-11 Greenhouse energy forecasting method based on hybrid optimization algorithm

Publications (2)

Publication Number Publication Date
CN103903068A true CN103903068A (en) 2014-07-02
CN103903068B CN103903068B (en) 2017-01-25

Family

ID=50994378

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201410144270.XA Active CN103903068B (en) 2014-04-11 2014-04-11 Greenhouse energy forecasting method based on hybrid optimization algorithm

Country Status (1)

Country Link
CN (1) CN103903068B (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107067035A (en) * 2017-04-17 2017-08-18 哈尔滨师范大学 The SVMs Study on wetland remote sensing method of coevolution algorithm optimization
CN107545100A (en) * 2017-07-31 2018-01-05 江苏大学 A kind of high-resolution Reduced Modeling Methods of venlo types greenhouse
CN110263907A (en) * 2019-07-01 2019-09-20 上海海事大学 Based on the ship short trouble diagnostic method for improving GA-PSO-BP

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2010051190A (en) * 2008-08-26 2010-03-11 Ehime Univ Apparatus and method for cooling local area of cultivated plant
CA2691228A1 (en) * 2010-01-27 2011-07-27 Trido Industries Inc. Control system for a chemical injection pump
CN102736596A (en) * 2012-06-11 2012-10-17 南京农业大学 Multi-scale greenhouse environment control system based on crop information fusion
CN202694113U (en) * 2012-06-29 2013-01-23 南京农业大学 Greenhouse environmental control system fused with crop information
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

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2010051190A (en) * 2008-08-26 2010-03-11 Ehime Univ Apparatus and method for cooling local area of cultivated plant
CA2691228A1 (en) * 2010-01-27 2011-07-27 Trido Industries Inc. Control system for a chemical injection pump
CN102736596A (en) * 2012-06-11 2012-10-17 南京农业大学 Multi-scale greenhouse environment control system based on crop information fusion
CN202694113U (en) * 2012-06-29 2013-01-23 南京农业大学 Greenhouse environmental control system fused with crop information
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

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107067035A (en) * 2017-04-17 2017-08-18 哈尔滨师范大学 The SVMs Study on wetland remote sensing method of coevolution algorithm optimization
CN107067035B (en) * 2017-04-17 2018-04-20 哈尔滨师范大学 The support vector machines Study on wetland remote sensing method of coevolution algorithm optimization
CN107545100A (en) * 2017-07-31 2018-01-05 江苏大学 A kind of high-resolution Reduced Modeling Methods of venlo types greenhouse
CN107545100B (en) * 2017-07-31 2020-05-05 江苏大学 High-resolution order-reduction modeling method for venlo type greenhouse environment
CN110263907A (en) * 2019-07-01 2019-09-20 上海海事大学 Based on the ship short trouble diagnostic method for improving GA-PSO-BP
CN110263907B (en) * 2019-07-01 2020-12-29 上海海事大学 Ship short-circuit fault diagnosis method based on improved GA-PSO-BP

Also Published As

Publication number Publication date
CN103903068B (en) 2017-01-25

Similar Documents

Publication Publication Date Title
Vanthoor et al. A methodology for model-based greenhouse design: Part 1, a greenhouse climate model for a broad range of designs and climates
Chen et al. Energy demand forecasting of the greenhouses using nonlinear models based on model optimized prediction method
Ghoulem et al. Greenhouse design and cooling technologies for sustainable food cultivation in hot climates: Review of current practice and future status
CN105117787B (en) A kind of model optimization prediction technique of greenhouse energy consumption
Villagrán et al. Transient CFD analysis of the natural ventilation of three types of greenhouses used for agricultural production in a tropical mountain climate
Ahamed et al. Development of a thermal model for simulation of supplemental heating requirements in Chinese-style solar greenhouses
Mohammadi et al. Application of dynamic model to predict some inside environment variables in a semi-solar greenhouse
He et al. The effect of vent openings on the microclimate inside multi-span greenhouses during summer and winter seasons
CN104133938A (en) Severe cold area office building external window geometric parameter optimization method based on heating energy consumption consideration
CN103903068A (en) Greenhouse energy forecasting method based on hybrid optimization algorithm
Yi et al. Site-specific optimal energy form generation based on hierarchical geometry relation
Ahamed et al. Modeling heating demands in a Chinese-style solar greenhouse using the transient building energy simulation model TRNSYS
CN106842914A (en) A kind of temperature control energy-saving processing method, apparatus and system
Munar et al. Study of natural ventilation in a Gothic multi-tunnel greenhouse designed to produce rose (Rosa spp.) in the high-Andean tropic
Baglivo et al. Complete greenhouse dynamic simulation tool to assess the crop thermal well-being and energy needs
Munar et al. Numerical evaluation of passive strategies for nocturnal climate optimization in a greenhouse designed for rose production (Rosa spp.)
Tan et al. An optimised window control strategy for naturally ventilated residential buildings in warm climates
Li et al. Evaluation of airflow pattern and thermal behavior of the arched greenhouses with designed roof ventilation scenarios using CFD simulation
Campen et al. Greenhouse cooling and heat recovery using fine wire heat exchangers in a closed pot plant greenhouse: design of an energy producing greenhouse
Baeza et al. Numerical simulation of the effect of different mulches on the heat storage capacity of a Mediterranean greenhouse soil
He et al. Effects of vent configuration and span number on greenhouse microclimate under summer conditions in Eastern China
Hong et al. Thermal behaviour of Trombe wall with venetian blind in summer and transition seasons
Belkadi et al. DESIGN AND IMPLEMENTATION OF FLC APPLIED TO A SMART GREENHOUSE
CN107765741A (en) A kind of greenhouse humiture decoupling control method and system
CN103077297B (en) A kind of Short-time-interval atmosphere ambient temperature prediction method

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