CN114117905A - Greenhouse crop irrigation method based on deep neural network - Google Patents
Greenhouse crop irrigation method based on deep neural network Download PDFInfo
- Publication number
- CN114117905A CN114117905A CN202111392679.XA CN202111392679A CN114117905A CN 114117905 A CN114117905 A CN 114117905A CN 202111392679 A CN202111392679 A CN 202111392679A CN 114117905 A CN114117905 A CN 114117905A
- Authority
- CN
- China
- Prior art keywords
- crop
- irrigation
- greenhouse
- neural network
- deep neural
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000003973 irrigation Methods 0.000 title claims abstract description 113
- 230000002262 irrigation Effects 0.000 title claims abstract description 113
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 27
- 238000000034 method Methods 0.000 title claims abstract description 22
- 230000005068 transpiration Effects 0.000 claims abstract description 46
- 238000011084 recovery Methods 0.000 claims abstract description 20
- 238000001514 detection method Methods 0.000 claims abstract description 19
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 34
- 208000005156 Dehydration Diseases 0.000 claims description 20
- 238000003062 neural network model Methods 0.000 claims description 12
- 230000000243 photosynthetic effect Effects 0.000 claims description 7
- 230000006735 deficit Effects 0.000 claims description 3
- 239000003621 irrigation water Substances 0.000 abstract description 2
- 241000196324 Embryophyta Species 0.000 description 11
- 240000003768 Solanum lycopersicum Species 0.000 description 8
- 235000007688 Lycopersicon esculentum Nutrition 0.000 description 7
- 238000005286 illumination Methods 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 3
- 230000002354 daily effect Effects 0.000 description 3
- 230000007613 environmental effect Effects 0.000 description 3
- 210000002569 neuron Anatomy 0.000 description 3
- 230000005855 radiation Effects 0.000 description 3
- 239000000758 substrate Substances 0.000 description 3
- 230000008020 evaporation Effects 0.000 description 2
- 238000001704 evaporation Methods 0.000 description 2
- 239000002689 soil Substances 0.000 description 2
- 239000002699 waste material Substances 0.000 description 2
- 235000016623 Fragaria vesca Nutrition 0.000 description 1
- 240000009088 Fragaria x ananassa Species 0.000 description 1
- 235000011363 Fragaria x ananassa Nutrition 0.000 description 1
- 241000276450 Hucho Species 0.000 description 1
- 241000607479 Yersinia pestis Species 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000001186 cumulative effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 230000003203 everyday effect Effects 0.000 description 1
- 239000003337 fertilizer Substances 0.000 description 1
- 238000003306 harvesting Methods 0.000 description 1
- 238000003064 k means clustering Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 241000894007 species Species 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- A—HUMAN NECESSITIES
- A01—AGRICULTURE; FORESTRY; ANIMAL HUSBANDRY; HUNTING; TRAPPING; FISHING
- A01G—HORTICULTURE; CULTIVATION OF VEGETABLES, FLOWERS, RICE, FRUIT, VINES, HOPS OR SEAWEED; FORESTRY; WATERING
- A01G9/00—Cultivation in receptacles, forcing-frames or greenhouses; Edging for beds, lawn or the like
- A01G9/24—Devices or systems for heating, ventilating, regulating temperature, illuminating, or watering, in greenhouses, forcing-frames, or the like
- A01G9/247—Watering arrangements
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/02—Agriculture; Fishing; Forestry; Mining
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A40/00—Adaptation technologies in agriculture, forestry, livestock or agroalimentary production
- Y02A40/10—Adaptation technologies in agriculture, forestry, livestock or agroalimentary production in agriculture
- Y02A40/25—Greenhouse technology, e.g. cooling systems therefor
Abstract
The invention provides a greenhouse crop irrigation method based on a deep neural network, which comprises the steps of firstly, constructing a crop transpiration rate prediction model, a greenhouse environment parameter prediction model, a crop moisture state detection model and a crop moisture stress recovery time prediction model based on the deep neural network; then inputting the predicted values of the crop planting date, the current time and the greenhouse environment parameter into a crop transpiration rate prediction model to obtain a crop transpiration rate predicted value, and integrating the time in the interval of the next irrigation time to obtain a crop transpiration amount predicted value; inputting the planting date, the current time, the greenhouse environment parameters, the RGB image of the crop canopy leaf, the depth image and the near-infrared image into a crop moisture state detection model to judge the moisture state of the crop, and irrigating according to the moisture state of the crop and the predicted value of the crop transpiration amount. The invention can dynamically adjust the irrigation time and the irrigation quantity, realize the accurate irrigation of crops and save the irrigation water.
Description
Technical Field
The invention belongs to the technical field of greenhouse crop irrigation, and particularly relates to a greenhouse crop irrigation method based on a deep neural network.
Background
Irrigation is the only source of moisture for greenhouse crops. Too much and too little water supply can affect the yield and quality of the crops to some extent. The root system of crops can be damaged due to excessive water supply, and the pest and disease damage and water waste are caused due to overhigh air humidity in the greenhouse; too little water supply will reduce the photosynthetic rate of the crop, resulting in reduced yield. The water demand of crops in different growth stages varies every day. The water demand of crops is small in the seedling stage, the water demand of crops is gradually increased along with the growth of the crops, and the water demand is reduced in the harvest period for controlling the quality. The changes of the temperature, the humidity and the illumination in the greenhouse at various times in a day also affect the water demand of crops, and the water demand of the crops is increased under the conditions of high temperature and high illumination intensity; the water demand of crops is reduced under the conditions of low temperature and low illumination.
Chinese patent (CN106258855A) discloses an intelligent irrigation system based on light radiation, which performs quantitative irrigation when the cumulative energy of the light radiation is greater than a predetermined critical energy. In 2018, hucho and the like, a sunlight greenhouse real-time accurate irrigation decision based on evapotranspiration and water balance is disclosed in agricultural engineering journal 34, 23, and irrigation is triggered when the total field evaporation amount is larger than the available water for crops in soil by taking days as a time scale, wherein the irrigation amount is equal to the total evapotranspiration amount since the last irrigation period. Chanswang et al, 2019, volume 50, journal addition of agricultural machinery, discloses a prediction model of daily transpiration of a single tomato plant, which takes variation of moisture content of a substrate, air temperature, air humidity and illumination intensity as input. Xulihong et al, in 2020, published tomato matrix cultivation irrigation model based on greenhouse environment and crop growth in the 10 th phase of volume 36 of agricultural engineering newspaper, modified the original expression of the Penman-Monteith model to remove the soil evaporation part, introduced TOMGRO model to simulate the growth of tomato canopy to correct the impedance parameter, and obtained a new transpiration model; and the influence of solar radiation on water consumption below the canopy of the crop is considered, and a daily irrigation quantity model of the single tomato is established. Li et al, 2020, Vol 51, No. 1, disclose research on strawberry irrigation strategy based on K-means clustering algorithm, and perform quantitative irrigation according to substrate water content change and daily average temperature interval. In the prior art, irrigation decision is mainly made by considering greenhouse environment parameters, substrate water content and the like, actual growth conditions of crops are not considered, prediction is only carried out by taking days as a unit, irrigation time and irrigation quantity cannot be dynamically adjusted according to the actual environment parameters and the actual water demand state of the crops on the same day, and the requirement for accurate irrigation of greenhouse crops is difficult to meet.
Disclosure of Invention
Aiming at the defects in the prior art, the invention provides a greenhouse crop irrigation method based on a deep neural network, which can detect the moisture state of crops in time, dynamically adjust the irrigation time and irrigation quantity according to the moisture state of the crops and environmental parameters, and realize accurate irrigation of the crops.
The present invention achieves the above-described object by the following technical means.
A greenhouse crop irrigation method based on a deep neural network specifically comprises the following steps:
collecting greenhouse environment parameters and RGB (red, green and blue) images, depth images and near-infrared images of crop canopy leaves every m minutes, inputting the greenhouse environment parameters and the RGB images into a crop moisture state detection model, and judging the moisture state of crops;
if the water state of the crops is irrigation shortage, immediately irrigating; and carrying out next crop water state detection after the crop water stress recovery time;
if the water state of the crops is insufficient after the irrigation within n hours, judging the water state of the crops through a crop water state detection model within n hours after the irrigation, and irrigating; wherein n is the irrigation interval.
Further, the crop moisture state detection model specifically comprises: and establishing a deep neural network model for detecting the moisture state of the crops by taking the planting date, the current time, the greenhouse environment parameters, the RGB image, the depth image and the near infrared image of the leaves of the canopy of the crops as input and the moisture state of the crops as output.
Further, the irrigation deficit is a crop moisture state with an irrigation level smaller than F (1-P), and the irrigation level with the highest net photosynthetic rate is defined as an irrigation reference and is recorded as F; the allowable relative deviation is defined as a threshold value, denoted as P.
Furthermore, the crop moisture state with the irrigation level larger than F (1+ P) is judged as the excessive irrigation, and the crop moisture state in the interval [ F (1-P), F (1+ P) ] is judged as the proper irrigation.
Further, the crop water stress recovery time is predicted according to a crop water stress recovery time prediction model, wherein the crop water stress recovery time prediction model specifically comprises: and (3) establishing a deep neural network model for predicting the crop water stress recovery time by taking the planting date, the current time, the greenhouse environment parameters and the irrigation quantity of the crops as input and the crop water stress recovery time as output.
Further, the irrigation amount is determined by the following formula:
wherein: q is the irrigation quantity of the individual plant, QpredAnd (3) the predicted value of the single plant transpiration in the next n hours, Q' is the single plant irrigation amount during the last irrigation, and ET is the single plant transpiration amount from the last irrigation to the current time.
Furthermore, the predicted value Q of the transpiration amount of the single plantpredThe acquisition method comprises the following steps: inputting greenhouse environment parameters of every minute in the first 24 hours, the whole-point weather forecast information in the last n hours and the set value of the control parameter of the greenhouse environment regulation and control system into a greenhouse environment parameter prediction model to obtain greenhouse environment prediction information in the last n hours; inputting the planting date, the current time and greenhouse environment prediction information of the crops into a crop transpiration rate prediction model, and calculating to obtain the crop transpiration rate; the integral calculation of the crop transpiration rate in the last n hoursObtaining a predicted value Q of the crop transpiration amountpred。
Furthermore, the crop transpiration rate prediction model specifically comprises: and (3) taking the planting date, the current time and the greenhouse environment parameters of the crops as the input of the model, and taking the crop transpiration rate as the output of the model to construct a deep neural network model for predicting the crop transpiration rate.
Furthermore, the greenhouse environment parameter prediction model specifically comprises: and taking the greenhouse environment parameters of the first 24 hours per minute, the whole-point weather forecast information in the last n hours and the control parameter set value of the greenhouse environment regulation and control system as inputs, taking the greenhouse environment parameters of every minute in the last n hours as outputs, and constructing a deep neural network model for greenhouse environment parameter prediction.
Further, m and n satisfy (n × 60) divisible by m.
The invention has the beneficial effects that: according to the invention, the moisture state of the crops can be judged in time for irrigation according to the field planting date, the current time, the greenhouse environment parameters, the RGB image, the depth image and the near-infrared image of the crop canopy leaves, the influence of greenhouse environment regulation on the greenhouse environment is considered, and then the irrigation quantity is dynamically adjusted according to the greenhouse environment parameters and the current time, so that the accurate irrigation of the crops is realized, the irrigation water can be saved, and the waste of water resources is reduced.
Drawings
FIG. 1 is a flow chart of a greenhouse crop irrigation method based on a deep neural network.
Detailed Description
The invention will be further described with reference to the following figures and specific examples, but the scope of the invention is not limited thereto.
The embodiment takes a greenhouse tomato irrigation method based on a deep neural network as an example, and further explains the implementation process.
S1, constructing a crop transpiration rate prediction model based on the deep neural network
Cultivating the crops required by establishing the crop transpiration rate prediction model, fully irrigating the crops, recording the planting date of the crops, and collecting the current date, the current time, the greenhouse environment parameters and the crop transpiration rate once every minute.
And (3) taking the planting date, the current time and the greenhouse environment parameters of the crops as the input of the model, and taking the crop transpiration rate as the output of the model to construct a deep neural network model for predicting the crop transpiration rate. In this example, the number of the tomato transpiration rate prediction model is 3, and the number of neurons in the hidden layer is 6. And training by using the acquired data to obtain a crop transpiration rate prediction model based on the deep neural network.
In this embodiment, greenhouse environmental parameters are collected by a greenhouse weather station, and the transpiration rate of the tomatoes is measured by a wrapped stem flow system.
S2, constructing a greenhouse environment parameter prediction model based on the deep neural network
The greenhouse environment regulation and control system automatically regulates, and records greenhouse environment parameters, current weather forecast and control parameter set values of the greenhouse environment regulation and control system once per minute. The time interval between two times of irrigation of the crop is defined as the irrigation interval time, which is recorded as n and is expressed in hours, and in the example, n is 1. And taking the greenhouse environment parameters of the first 24 hours per minute, the whole-point weather forecast information in the last n hours and the control parameter set value of the greenhouse environment regulation and control system as inputs, taking the greenhouse environment parameters of every minute in the last n hours as outputs, and constructing a deep neural network model for greenhouse environment parameter prediction. In this embodiment, the number of layers of the greenhouse environment parameter prediction model is 3, and the number of neurons in the hidden layer is 45. And training by using the recorded greenhouse environment parameters, weather forecast information and control parameter set values of the greenhouse environment regulation and control system to obtain a greenhouse environment parameter prediction model based on the deep neural network.
S3, establishing a crop moisture state detection model based on a deep neural network
Cultivating the crops required for establishing the crop moisture state detection model, and setting different irrigation levels to enable the crops to be in different moisture states. Collecting the field planting date, the current time, greenhouse environment parameters, RGB images of crop canopy leaves, depth images, near-infrared images and net photosynthetic rate once per minute; defining the irrigation level with the highest net photosynthetic rate as an irrigation reference, and recording the irrigation level as F; defining the allowable relative deviation as a threshold value, which is denoted as P, and in the example, P is 5%; the crop moisture state with the irrigation level larger than F (1+ P) is judged as excessive irrigation, the crop moisture state in the interval [ F (1-P), F (1+ P) ] is judged as proper irrigation, and the crop moisture state smaller than F (1-P) is judged as insufficient irrigation.
And establishing a deep neural network model for detecting the moisture state of the crops by taking the planting date, the current time, the greenhouse environment parameters, the RGB image, the depth image and the near infrared image of the leaves of the canopy of the crops as input and the moisture state of the crops as output. And training by using the collected greenhouse environment parameters and the RGB image, the depth image, the near infrared image and the crop moisture state of the crop canopy leaf to obtain a crop moisture state detection model based on the deep neural network.
The RGB images, depth images and near-infrared images of the crop canopy leaves are collected by a D435i depth camera of RealSense.
The net photosynthetic rate is measured using a photosynthetic apparatus.
S4, establishing a crop water stress recovery time prediction model based on a deep neural network
Cultivating crops required by establishing a crop water stress recovery time prediction model, and inputting the field planting date, the current time, greenhouse environment parameters, RGB images, depth images and near-infrared images of crop canopy leaves into a crop water state detection model to obtain the crop water state. When the crops are in the irrigation shortage state, the crops are fully irrigated, and the planting date, the current time, the greenhouse environment parameters and the irrigation amount of the crops are recorded. After that, the crop water state is detected once every minute until the crop water state is no longer irrigation deficit, and the elapsed time is recorded as the crop water stress recovery time.
And (3) establishing a deep neural network model for predicting the crop water stress recovery time by taking the planting date, the current time, the greenhouse environment parameters and the irrigation quantity of the crops as input and the crop water stress recovery time as output. In this embodiment, the number of the tomato water stress time prediction model layers is 3, and the number of neurons in the hidden layer is 6. And training by using the acquired data to obtain a crop water stress recovery time prediction model based on the deep neural network.
S5, making a greenhouse crop irrigation strategy
Greenhouse environmental parameters were collected every one minute during the day. Carrying out first irrigation after sunrise time of the place of the greenhouse, and calculating irrigation quantity according to a formula (1) that the moisture state is an appropriate irrigation state;
the time interval between two crop moisture state detections is defined as the crop moisture state detection time interval, which is recorded as m, and the unit is minutes, and in this embodiment, m is 10. And (3) collecting greenhouse environment parameters and RGB (red, green and blue) images, depth images and near-infrared images of crop canopy leaves every m minutes, and inputting the information into a crop moisture state detection model to judge the moisture state of the crops. And if the water state of the crops is insufficient for irrigation, immediately irrigating, calculating the irrigation quantity by the formula (1), and detecting the water state of the crops next time after the water stress recovery time of the crops.
If the moisture state of the crops is insufficient for irrigation (namely excessive irrigation or proper irrigation) within n hours after irrigation, judging the moisture state of the crops through a crop moisture state detection model and starting an irrigation fertilizer applicator to irrigate n hours after irrigation, wherein the irrigation quantity is calculated by the formula (1):
wherein Q is the irrigation quantity of a single plant, and the unit is as follows: mL/plant;
Qpredpredicted values of the transpiration amount of each plant in the next n hours are as follows: mL/plant;
q' is the irrigation quantity of each plant during the last irrigation, and the unit is as follows: mL/plant;
ET is the transpiration of a single plant in the time period from the last irrigation to the current time, and the unit is as follows: the calculation method comprises the steps of firstly calculating the crop transpiration rate in the time period from the last irrigation to the current time according to a crop transpiration rate prediction model, and then integrating the crop transpiration rate in the time period to obtain a value, namely the crop transpiration amount in the time period.
The predicted value Q of the transpiration amount of the single plantpredIs determined by the following method: inputting the greenhouse environment parameters of the first 24 hours per minute, the whole-point weather forecast information in the last n hours and the set values of the control parameters of the greenhouse environment regulation and control system into the greenhouse environment parameter prediction model established in S2 to obtain the greenhouse environment prediction information in the last n hours; inputting the planting date, the current time and the greenhouse environment prediction information of the crops into the crop transpiration rate prediction model established in the S1 to calculate the crop transpiration rate; and performing integral calculation on the crop transpiration rate in the next n hours to obtain a crop transpiration amount predicted value Qpred。
The crop moisture status detection time interval of m minutes and the irrigation time interval of n hours are adjusted according to the crop species, and (n × 60) can be evenly divided by m.
And S6, irrigating the crops according to the greenhouse crop irrigation strategy. In this example, the fertigation machine irrigates according to a greenhouse crop irrigation strategy.
The present invention is not limited to the above-described embodiments, and any obvious improvements, substitutions or modifications can be made by those skilled in the art without departing from the spirit of the present invention.
Claims (10)
1. A greenhouse crop irrigation method based on a deep neural network is characterized in that:
collecting greenhouse environment parameters and RGB (red, green and blue) images, depth images and near-infrared images of crop canopy leaves every m minutes, inputting the greenhouse environment parameters and the RGB images into a crop moisture state detection model, and judging the moisture state of crops;
if the water state of the crops is irrigation shortage, immediately irrigating; and carrying out next crop water state detection after the crop water stress recovery time;
if the water state of the crops is insufficient after the irrigation within n hours, judging the water state of the crops through a crop water state detection model within n hours after the irrigation, and irrigating; wherein n is the irrigation interval.
2. The deep neural network-based greenhouse crop irrigation method as claimed in claim 1, wherein the crop moisture state detection model is specifically: and establishing a deep neural network model for detecting the moisture state of the crops by taking the planting date, the current time, the greenhouse environment parameters, the RGB image, the depth image and the near infrared image of the leaves of the canopy of the crops as input and the moisture state of the crops as output.
3. The deep neural network-based greenhouse crop irrigation method as claimed in claim 1, wherein the irrigation deficit is a crop moisture state with irrigation level less than F (1-P), and the irrigation level with the highest net photosynthetic rate is defined as an irrigation reference, denoted as F; the allowable relative deviation is defined as a threshold value, denoted as P.
4. The deep neural network-based greenhouse crop irrigation method as claimed in claim 3, wherein the crop moisture status with irrigation level greater than F (1+ P) is judged as irrigation excess, and the crop moisture status in the interval [ F (1-P), F (1+ P) ] is judged as irrigation proper.
5. The deep neural network-based greenhouse crop irrigation method as claimed in claim 1, wherein the crop water stress recovery time is predicted according to a crop water stress recovery time prediction model, and the crop water stress recovery time prediction model is specifically: and (3) establishing a deep neural network model for predicting the crop water stress recovery time by taking the planting date, the current time, the greenhouse environment parameters and the irrigation quantity of the crops as input and the crop water stress recovery time as output.
6. The deep neural network-based greenhouse crop irrigation method of claim 4, wherein the irrigation amount is determined by the following formula:
wherein: q is the irrigation quantity of the individual plant, QpredAnd (3) the predicted value of the single plant transpiration in the next n hours, Q' is the single plant irrigation amount during the last irrigation, and ET is the single plant transpiration amount from the last irrigation to the current time.
7. The deep neural network-based greenhouse crop irrigation method as claimed in claim 6, wherein the single plant transpiration predicted value QpredThe acquisition method comprises the following steps: inputting greenhouse environment parameters of every minute in the first 24 hours, the whole-point weather forecast information in the last n hours and the set value of the control parameter of the greenhouse environment regulation and control system into a greenhouse environment parameter prediction model to obtain greenhouse environment prediction information in the last n hours; inputting the planting date, the current time and greenhouse environment prediction information of the crops into a crop transpiration rate prediction model, and calculating to obtain the crop transpiration rate; the crop transpiration rate is integrated in the last n hours to obtain a crop transpiration amount predicted value Qpred。
8. The deep neural network-based greenhouse crop irrigation method as claimed in claim 7, wherein the crop transpiration rate prediction model is specifically: and (3) taking the planting date, the current time and the greenhouse environment parameters of the crops as the input of the model, and taking the crop transpiration rate as the output of the model to construct a deep neural network model for predicting the crop transpiration rate.
9. The deep neural network-based greenhouse crop irrigation method as claimed in claim 7, wherein the greenhouse environment parameter prediction model is specifically: and taking the greenhouse environment parameters of the first 24 hours per minute, the whole-point weather forecast information in the last n hours and the control parameter set value of the greenhouse environment regulation and control system as inputs, taking the greenhouse environment parameters of every minute in the last n hours as outputs, and constructing a deep neural network model for greenhouse environment parameter prediction.
10. The deep neural network-based greenhouse crop irrigation method as claimed in claim 1, wherein m and n satisfy (n x 60) divisible by m.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111392679.XA CN114117905A (en) | 2021-11-23 | 2021-11-23 | Greenhouse crop irrigation method based on deep neural network |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111392679.XA CN114117905A (en) | 2021-11-23 | 2021-11-23 | Greenhouse crop irrigation method based on deep neural network |
Publications (1)
Publication Number | Publication Date |
---|---|
CN114117905A true CN114117905A (en) | 2022-03-01 |
Family
ID=80440314
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202111392679.XA Pending CN114117905A (en) | 2021-11-23 | 2021-11-23 | Greenhouse crop irrigation method based on deep neural network |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114117905A (en) |
-
2021
- 2021-11-23 CN CN202111392679.XA patent/CN114117905A/en active Pending
Similar Documents
Publication | Publication Date | Title |
---|---|---|
RU2688234C1 (en) | Smart growing control method and smart device for growing | |
CN109191074A (en) | Wisdom orchard planting management system | |
CN106951017B (en) | The wisdom Greenhouse System of crop regulation-control model driving regulates and controls method | |
CN108781926B (en) | Greenhouse irrigation system and method based on neural network prediction | |
CN102823402B (en) | Efficient and environment friendly irrigation method for rices in cold region | |
CN105494033B (en) | A kind of intelligent water-saving irrigation method based on crop demand | |
CN108617355A (en) | A kind of cluster greenhouse irrigation decision-making technique and system | |
KR20130041703A (en) | The inside green house ginseng grawing method and system by variable artificial light source | |
CN111640038B (en) | Rice crop coefficient calculation method and rice irrigation system | |
CN110447509B (en) | Nutrient solution irrigation control system and method for plant matrix cultivation | |
CN109116897A (en) | A kind of agricultural planting ambient intelligence adjusting method based on big data | |
CN112167035B (en) | Hydroponic leaf vegetable production management method and system | |
CN112314414B (en) | Automatic control method for tide irrigation plug crops | |
CN113597941A (en) | Greenhouse intelligent environment regulation and control system and device | |
CN111176238B (en) | AIPA intelligent decision-making type precise agricultural system | |
CN105993720B (en) | Simulation calculation method for irrigation quantity of matrix bag-cultured crops in sunlight greenhouse | |
CN109644767B (en) | Method for cultivating oil peony | |
CN116755485A (en) | Greenhouse regulation and control method, device, system, equipment and storage medium | |
CN114117905A (en) | Greenhouse crop irrigation method based on deep neural network | |
Morgan | Daily Light Integral (DLI) and greenhouse tomato production | |
CN115349432A (en) | Intelligent irrigation method and system based on optical radiation and meteorological prediction | |
Giacomelli et al. | Design of a single truss tomato production system (STTPS) | |
JP6785902B2 (en) | Crop activity index-based facility Horticulture complex environmental control system and method | |
CN214708748U (en) | Management and control system for accurate nutrition irrigation of crops | |
Aharari et al. | Development of IoT-based smart agriculture monitoring system for red radish plants production |
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 |