CN115804334A - Farmland irrigation intelligent decision system based on digital twins - Google Patents
Farmland irrigation intelligent decision system based on digital twins Download PDFInfo
- Publication number
- CN115804334A CN115804334A CN202211508391.9A CN202211508391A CN115804334A CN 115804334 A CN115804334 A CN 115804334A CN 202211508391 A CN202211508391 A CN 202211508391A CN 115804334 A CN115804334 A CN 115804334A
- Authority
- CN
- China
- Prior art keywords
- irrigation
- data
- layer
- intelligent
- decision
- 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
Images
Classifications
-
- 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
- Y02A40/22—Improving land use; Improving water use or availability; Controlling erosion
Abstract
The invention provides a farmland irrigation intelligent decision system based on digital twinning, which comprises a physical layer, a data acquisition layer, a twinning model layer, a functional layer and an application layer, wherein the physical layer is a physical layer; the physical layer comprises sprinkling irrigation equipment and hardware equipment; the data acquisition layer is used for acquiring farmland information and equipment running states; the twin model layer is used for twin data fusion and construction of a digital twin model; the functional layer comprises a visual real-time monitoring platform, a precise irrigation decision-making system and an intelligent irrigation control system, wherein the precise irrigation decision-making system makes an intelligent decision on irrigation quantity and irrigation time by depending on an irrigation decision-making algorithm on the basis of air temperature and humidity, wind speed, sunshine hours and pipeline flow data, and the intelligent irrigation control system forms a command for remotely controlling irrigation equipment to perform irrigation action according to the actual required irrigation time obtained by the irrigation decision-making system; the intelligent farmland scene monitoring system realizes the functions of visual monitoring, intelligent early warning and irrigation of the intelligent farmland scene in a virtual space.
Description
Technical Field
The invention relates to the field of intelligent agriculture, in particular to a farmland irrigation intelligent decision system based on digital twins.
Background
At present, in wisdom irrigation field, multiple water-saving irrigation technique combines together with emerging information technology, and the irrigation process begins to develop to automation, informationization, teleization, visualization, intellectuality. How to utilize the information data of gathering in real time in the farmland, establish a multi-functional, collect automatic, visual, long-range, intelligent farmland irrigation wisdom decision-making system who integrates, be the urgent affairs of present wisdom agricultural development.
At present, most of farmland irrigation is still in a manual control management stage, data of farmland equipment, crops and the like are manually collected or manually input at regular intervals, the data volume is small, the real-time performance is poor, and a user can hardly visually and timely acquire the environment state of the farmland and the growth condition of the crops; in addition, most of farmland irrigation equipment is manually switched on and off, a mature and reliable control strategy is lacked, and automatic control is difficult to realize. Although a part of farmlands adopt the advanced internet of things technology, the real-time collection of data such as the farmland environment, the soil moisture content of crops and the like is realized, and users can also receive corresponding data in time. However, the functions of the system in this mode are limited to real-time monitoring of data, which is too single, and the functions of intelligent decision making, control and the like are still weak.
Disclosure of Invention
Aiming at the defects in the prior art, the invention provides a farmland irrigation intelligent decision-making system based on a digital twin, solves the technical problems that the existing intelligent farmland irrigation system is single in function, non-visual in data, poor in real-time performance and incapable of effectively interconnecting and sharing data information, and realizes the visual monitoring, intelligent early warning and irrigation functions of intelligent farmland scenes in a virtual space.
The present invention achieves the above-described object by the following means.
A farmland irrigation intelligent decision system based on digital twins comprises a physical layer, a data acquisition layer, a twins model layer, a functional layer and an application layer;
the physical layer comprises sprinkling irrigation equipment and hardware equipment, wherein the hardware equipment comprises a soil temperature and humidity sensor, a camera, an on-site meteorological station, an information acquisition gateway and a controller;
the data acquisition layer is used for acquiring farmland information and equipment running states and comprises a data acquisition module and a remote communication module;
the twin model layer is used for twin data fusion and construction of a digital twin model;
the functional layer comprises a visual real-time monitoring platform, a precise irrigation decision-making system and an intelligent irrigation control system, wherein the precise irrigation decision-making system makes an intelligent decision on irrigation quantity and irrigation time based on air temperature and humidity, wind speed, sunshine hours and pipeline flow data and by depending on an irrigation decision-making algorithm, and the intelligent irrigation control system forms an instruction for remotely controlling irrigation equipment to perform irrigation action according to actual required irrigation time obtained by the irrigation decision-making system;
the system application layer develops an interface on a UI interface, and three-dimensional display, equipment maintenance, irrigation decision, intelligent early warning, real-time monitoring, chart statistics and remote control are realized.
Furthermore, the data acquisition layer comprises an acquisition module and a remote communication module, wherein the data acquisition module is used for acquiring field sensor data of different protocols and supporting the access of analog quantity, digital quantity and RS485 data; the data acquisition module and the remote communication module are communicated through a Lora wireless network, the remote communication module packs data of each data acquisition device, then processes the data into MQTT protocol data, and data transmission is realized through a 4G/5G, ethernet or WiFi and a cloud platform.
Furthermore, the twin data fusion is carried out on a time scale and a space scale, and negligence errors of the data are eliminated by adopting a distribution diagram method aiming at the temperature, humidity and illumination intensity data parameters on the time scale; on the spatial scale, the temperature and humidity, pipeline pressure and flow data are fused by using a weighted fusion algorithm to obtain accurate parameters, and then the fused data are input into a digital twin model.
Further, the distribution diagram method comprises the following specific steps:
s1: sequencing n measurement results measured by m sensors in a farmland from small to large to obtain a measurement sequence: x = (X) 1 ，x 2 ，…，x n )
S2: when n is an even number, a median value x is defined N Is composed of
When n is odd, a median value x is defined N Comprises the following steps:
s3: upper quartile F On the upper part Is [ x ] N ，x n ]Median, lower quartile F in interval Lower part Is [ x ] 1 ，x N ]The dispersion of the median and quartile over the interval is dF = F On the upper part -F Lower part ；
S4, setting the effective judgment interval to be F Lower part –dF，F On the upper part +dF]If the data is in the interval, the data is judged to be valid, and if the data is out of the interval, the data is judged to be negligent error and is rejected.
Further, the constructing of the digital twin model comprises the following steps:
step T1: oblique photography pictures of the farmland by using an unmanned aerial vehicle, and combining global digital elevation data to synthesize and process the pictures by using a Dajiang intelligent map software tool, so as to realize the construction of a three-dimensional scene of the farmland;
step T2: establishing three-dimensional models of a plurality of physical entities in an intelligent farmland scene by using a three-dimensional modeling tool, and solving structural parameters, geometric parameters, material parameters, state parameters and boundary conditions of the three-dimensional models by using a finite element analysis method;
step T3: rendering the model by using a 3D Max three-dimensional rendering tool according to the three-dimensional model obtained in the step T2, adding materials, and optimizing the edge part of the model;
and step T4: and (3) introducing the model rendered in the step T3 into a three-dimensional scene through three-dimensional visualization management software TerraExplorer, giving physical, geometric, behavioral and regular dynamic attributes to the digital twin model, and dynamically reflecting various running states of the entity irrigation equipment.
Furthermore, the visual real-time monitoring platform is used for acquiring environmental parameter information and soil temperature, humidity and soil moisture information acquired by the weather station and transmitting the acquired information to the precision irrigation decision system.
Furthermore, the intelligent irrigation decision system obtains the needed irrigation quantity of the crops through an irrigation decision algorithm according to the collected data information of the weather station, the temperature and humidity of the crop soil, the wind speed, the sunshine hours and the pipeline flow, and then obtains the actually needed irrigation time by combining the pipeline flow to form an irrigation decision control strategy.
Further, the irrigation decision algorithm comprises:
step A1: firstly, calculating the reference crop water demand ET according to a Penman-Monteith correction formula (1) 0 ；
Wherein, ET 0 : reference crop evapotranspiration amount, mm/d;
t: calculating the average air temperature 2m away from the ground within a time period;
Δ: the slope of the plant rising pressure curve, kPa/DEG C;
R n : net solar radiation, MJ/(m) 2 ·d)；
G: soil heat flux, MJ/(m) 2 ·d)；
γ: monitored dry and wet constants, kPa/deg.C;
e s : saturated water gas pressure, kPa; e.g. of a cylinder a : actual water gas pressure, kPa;
u 2 : average wind speed at 2m height from the ground, m/s;
step A2: calculating to obtain the actual water demand Etc of the crops by combining the water storage coefficient Kc of the growth stage of the crops, wherein the Kc refers to the FAO recommended value and is set according to the growth stage of the crops;
ET c ＝K c ET 0 (2)
a3: calculating the irrigation quantity of a unit area according to the temperature and humidity of air at a position 2m away from the ground, the wind speed and the sunshine hours, calculating the total irrigation quantity according to the irrigation quantity of the unit area and the land area, and finally calculating the irrigation duration according to the collected pipeline flow;
and S4, sending the irrigation quantity and the irrigation time to a system control module through a remote control.
Further, the irrigation control strategy adopts a soil moisture content prediction model to predict the soil humidity of the next time point, determines the irrigation strategy by comparing the soil humidity of the next time point with the soil humidity of the most suitable growth interval, reduces the dimension of the acquired data set by using PCA, and respectively establishes RBF prediction models of the soil moisture content by using the data sets before and after the dimension reduction; calling a newrbe function to design a radial basis network, and setting the extension speed spread to be 30.
Further, the method for establishing the RBF soil water content prediction model comprises the following steps:
b1: determining a weighting factor W1 from the input layer to the radial base layer, the weighting factor W1= R T ；
B2: solving for the bias b1 from the input layer to the radial base layer by the formula b1= sqrt (-log (.5))/spread;
b3: calculating The distance between each sample and The radial basis neuron according to The formula of The dist function, wherein | | x-W1| |;
b4: calculating a parameter n by a formula n = | | x-W1| | | b1;
b5: repeating the steps B1-B4, and after finishing, performing the step B6;
b6: by the formula Transfer function:
calculating an output result A {1} of the first layer;
b7: (ii) linear expressions [ W {2,1} b, face 2} ] [ A {1}; ones (1, q) ] = T solves for the weight coefficient W2 and the offset b2 of the second layer;
s8: finishing;
wherein, the input is R = [ T = c ,T air ,RH,R n ,V]The output is T = [ theta ]]
Where R is the input vector, T is the output vector, Q is the number of input vectors, A {1} is the output result of the first layer, b1 is the deviation of the input layer to the radial base layer, W1 is the weight of the input layer to the radial base layer, b2 is the deviation of the radial base layer to the output layer, and W2 is the weight of the radial base layer to the output layer.
The invention has the beneficial effects that:
according to the invention, through building a meteorological station and a wireless sensor network, accurate and real-time acquisition of data such as farmland meteorological environment, plant soil moisture content, equipment state and the like is realized; through data fusion processing, screening and optimization of multi-source heterogeneous data are realized, and the accuracy and reliability of the data are improved; by constructing the digital twin body, the mapping from the physical farm to the virtual farm is realized, and the data interconnection and sharing between the digital twin body and the physical entity are ensured. The intelligent farmland irrigation system can realize functions of data visualization, abnormal alarm, intelligent irrigation and the like of an intelligent farmland, and can greatly save manpower and material resources and save water resources.
Drawings
FIG. 1 is a diagram of a physical system of a digital twin-based intelligent decision system for field irrigation according to an embodiment of the present invention;
FIG. 2 is a block diagram of an embodiment of the present invention illustrating an overall architecture of a digital twin-based intelligent decision-making system for field irrigation;
FIG. 3 is a flow chart of a digital twin construction according to an embodiment of the present invention;
FIG. 4 is a diagram of a system control module according to an embodiment of the present invention;
FIG. 5 is a diagram of a system power circuit according to an embodiment of the present invention
Fig. 6 is a flow chart of an irrigation control strategy according to an embodiment of the present invention.
Detailed Description
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in detail below. In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention. This invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein, but rather should be construed as broadly as the present invention is capable of modification in various respects, all without departing from the spirit and scope of the present invention.
Referring to fig. 1 to 6, an intelligent decision system for digital twin-based field irrigation according to an embodiment of the present invention includes a physical layer, a data acquisition layer, a twin model layer, a functional layer, and an application layer.
Specifically, as shown in fig. 2, the physical layer includes a water source 1, a filter 2, a water pump 3, a pressure sensor 4, a flow sensor 5, a pressure reducing valve 6, an electromagnetic valve 7, a sprinkling irrigation nozzle 8, a soil humidity sensor 9, and a field weather station 10. Water in the water source 1 is carried out the filter 2 and is handled the back, is carried out by water pump 3, transmits to each irrigate the field, evenly sprays by sprinkling irrigation nozzle 8 and irrigates the operation.
Further, a water pump controller is integrated in the water pump 3 and used for remotely controlling the operation of the water pump controller; the vibration sensor, the temperature sensor, the flow sensor, the pressure sensor, the water level sensor and the like are used for monitoring the running state of the water level sensor. The pressure sensor 4, flow sensor 5, soil moisture sensor 9 and field weather station 10 all support Lora network communication.
The data acquisition layer is used for acquiring farmland information and equipment running states, and comprises a data acquisition unit, an edge gateway and an Internet of things cloud platform, wherein the data acquisition unit is used for acquiring field sensor data of different protocols, supports the access of analog quantity, digital quantity and RS485 data, and realizes the real-time acquisition of farmland environment data, soil moisture content data and equipment running states, and the data acquisition layer comprises data types such as soil temperature and humidity, environment temperature and humidity, illumination intensity and equipment running parameters. The data acquisition devices and the edge gateway are communicated through a Lora wireless network, the edge gateway packages data of the data acquisition devices, then processes the data into MQTT protocol data, and data transmission is achieved through 4G/5G, ethernet or WiFi and a cloud platform.
The twin model layer is used for twin data fusion and construction of a digital twin model; the twin data fusion is carried out on a time scale and a space scale, and negligence errors of the data are eliminated by adopting a distribution diagram method aiming at data parameters such as temperature, humidity and the like on the time scale; on the spatial scale, data such as temperature and humidity, pipeline pressure, flow and the like are fused by using a weighted fusion algorithm to obtain accurate parameters, and then the fused data are input into a digital twin model.
The distribution diagram method comprises the following specific steps:
s1: sequencing n measurement results measured by m sensors in a farmland from small to large to obtain a measurement sequence: x = (X) 1 ，x 2 ，…，x n )
S2: when n is an even number, a median value x is defined N Is composed of
When n is odd, define median x N Comprises the following steps:
s3: upper quartile F Upper part of Is [ x ] N ，x n ]Median, lower quartile F over interval Lower part Is [ x ] 1 ，x N ]The dispersion of the median and quartile in the interval is dF = F Upper part of -F Lower part ；
S4, setting the effective judgment interval to be F Lower part –dF，F Upper part of +dF]If the data is in the interval, the data is judged to be valid, and if the data is out of the interval, the data is judged to be negligent error and is rejected.
In this example, taking soil humidity collected by field sensors as an example, 2 sensors each continuously collected 8 soil humidity data at a frequency of 15 minutes/time as shown in the following table.
Calculating to obtain a median value x N =39.8, arithmetic mean T m =40, upper quartile F Upper part of =40.3, lower quartile F Lower part =39.5, dispersion dF of quartile =0.8, and effective determination interval is [38.7, 41.1 ]]Then, the three data of 38.4, 44.3 and 37.3 are eliminated.
Variance (delta) of each sensor obtained from previous data acquisition 2 ) Value, of the first sensor
Of a second sensor
Mean square error δ =0.172 was calculated for the 5 data after culling. Using a weighted fusion formula, W 1 ＝0.4584，W 2 =0.5416, fused humidity data S =39.956. The data fusion operation is finished, and the accuracy is high.
Further, the specific steps of constructing the digital twin model are as follows:
step T1: oblique photography pictures of a farmland by an unmanned aerial vehicle are utilized, and a global digital elevation Data (DEM) is combined by a Dajiang intelligent map software tool to synthesize and process the pictures, so that a three-dimensional scene of the farmland is constructed;
step T2: establishing a three-dimensional model of a plurality of physical entities (sensors, equipment and the like) in an intelligent farmland scene by using a three-dimensional modeling tool, and solving structural parameters, geometric parameters, material parameters, state parameters and boundary conditions of the three-dimensional model by using a finite element analysis method;
and step T3: rendering and adding materials to the model by using a 3D Max three-dimensional rendering tool according to the three-dimensional model obtained in the step T2, and optimizing the edge part of the model;
and step T4: and (4) introducing the model rendered in the step T3 into a three-dimensional scene through three-dimensional visualization management software Terra Explorer, endowing the digital twin model with physical, geometric, behavioral and regular dynamic attributes, dynamically reflecting various running states of entity irrigation equipment, and realizing visualization modeling of a digital twin body and visualization display of a virtual space intelligent farmland application scene.
The functional layer comprises a visual real-time monitoring platform, a precision irrigation decision-making system and an intelligent irrigation control system; the intelligent soil moisture content prediction method aims to monitor information data of a farmland and the running state of equipment in real time through a digital twin visual front end, realize prediction of soil moisture content by embedding a PCA-RBF algorithm model, and intelligently decide irrigation quantity and irrigation time, and instructions generated by decision can be input into a Microcontroller (MCU) through a 4G/5G wireless transmission module, so that automatic control of irrigation equipment is realized. Meanwhile, information interaction between the physical farmland and the digital twin platform is realized, and field data, equipment state, alarm information, decision results and the like are presented in the form of data and diagrams. The user can input different control commands according to the situation to remotely adjust the running state of the equipment.
The visual real-time monitoring platform is used for acquiring environmental parameter information and soil temperature, humidity and soil moisture information acquired by the weather station and transmitting the acquired information to the precision irrigation decision-making system. The intelligent irrigation decision-making system obtains the irrigation quantity needed by crops through an irrigation decision-making algorithm according to the collected data information of the weather station, the temperature and humidity of the soil of the crops, the wind speed, the sunshine hours and the pipeline flow, and then obtains the actual needed irrigation time by combining the pipeline flow to form an irrigation decision-making control strategy. And the intelligent irrigation control system forms an instruction for remotely controlling irrigation equipment to irrigate according to the actual required irrigation time obtained by the irrigation decision system.
Further, an irrigation decision algorithm is based on the reference crop transpiration amount ET 0 In combination with the water demand coefficient K of the growing stage of the crop c Obtaining the actual water demand ET of the crops c And then, calculating the total irrigation quantity according to the irrigation quantity per unit area and the land area, and combining the prediction result of the soil humidity at the next moment to realize the intelligent decision of the current irrigation quantity and the irrigation duration. The method specifically comprises the following steps:
step A1: firstly, calculating the reference crop water demand ET according to a Penman-Monteith correction formula (1) 0 ；
Wherein, ET 0 : reference crop evapotranspiration amount, mm/d;
t: calculating the average air temperature 2m away from the ground within a time period;
Δ: the slope of the plant transpiration pressure curve, kPa/DEG C;
R n : net solar radiation, MJ/(m) 2 ·d)；
G: soil heat flux, MJ/(m) 2 ·d)；
γ: monitored dry and wet constants, kPa/deg.C;
e s : saturated water gas pressure, kPa; e.g. of a cylinder a : actual water gas pressure, kPa;
u 2 : average wind speed at 2m height from the ground, m/s;
step A2: calculating to obtain the actual water demand Etc of the crops by combining the water storage coefficient Kc of the growth stage of the crops, wherein the Kc refers to the FAO recommended value and is set according to the growth stage of the crops;
ET c ＝K c ET 0 (2)
a3: calculating the irrigation quantity per unit area according to the air temperature and humidity, the air speed and the sunshine hours at a position 2m away from the ground, calculating the total irrigation quantity according to the irrigation quantity per unit area and the land area, and finally calculating the irrigation duration according to the collected pipeline flow;
and S4, remotely sending information such as irrigation quantity, irrigation time and the like to a system control module.
Furthermore, the hardware structure diagram of the intelligent irrigation control system is shown in fig. 4, the power supply module supplies power to a Microcontroller (MCU) and a 4G/5G wireless communication module, and the MCU controls the relay switch according to the flow to drive the solenoid valve to work.
The irrigation control strategy flow is as shown in fig. 6, and is characterized in that a Microcontroller (MCU) is used as a core, and after receiving an instruction wirelessly, a relay module is controlled to drive the rotation speed of a motor or a solenoid valve switch to execute irrigation action, so as to judge whether the predicted irrigation water quantity is reached or not, and automatically determine the on-off state of irrigation equipment.
The irrigation control strategy adopts a soil moisture content prediction model to predict the soil moisture content at the next time point, determines the irrigation strategy by comparing the soil moisture content at the next time with the soil moisture content at the most suitable growth interval, reduces the dimension of the acquired data set by using PCA, and respectively establishes RBF prediction models of the soil moisture content by using the data sets before and after the dimension reduction; calling a newrbe function to design a radial basis network, and setting the extension speed spread to be 30.
The method for establishing the RBF soil water content prediction model comprises the following steps:
b1: determining a weighting factor W1 from the input layer to the radial base layer, the weighting factor W1= R T ；
B2: solving for the bias b1 from the input layer to the radial base layer by the formula b1= sqrt (-log (.5))/spread;
b3: calculating The distance between each sample and The radial basis neuron according to The formula of The dist function, wherein | | x-W1| |;
b4: calculating a parameter n by a formula n = | | x-W1| | | b1;
b5: repeating the steps B1-B4, and after finishing, performing B6;
b6: by the formula Transfer function:
calculating an output result A {1} of the first layer;
b7: (ii) linear expressions [ W {2,1} b, face 2} ] [ A {1}; ones (1, q) ] = T solves for the weight coefficient W2 and the offset b2 of the second layer;
s8: ending;
wherein the input is R = [ T = c ,T air ,RH,R n ,V]The output is T = [ theta ]]
Where R is the input vector, T is the output vector, Q is the number of input vectors, A {1} is the output result of the first layer, b1 is the deviation of the input layer to the radial base layer, W1 is the weight of the input layer to the radial base layer, b2 is the deviation of the radial base layer to the output layer, and W2 is the weight of the radial base layer to the output layer.
The system application layer develops an interface on a UI interface, and three-dimensional display, equipment maintenance, irrigation decision, intelligent early warning, real-time monitoring, chart statistics and remote control are realized. In order to complete the design of the human-computer interaction interface, the visualization interface is designed and arranged by utilizing ThingJS software. Twin data is accessed through the software built-in interface, and data docking forms such as Ajax, JSONP, webSocket and the like are supported. In addition, various visual charts and keys can be added on the basis of a three-dimensional scene, and the visual charts correspond to applications such as three-dimensional display, equipment maintenance, irrigation decision, intelligent early warning, real-time monitoring, chart statistics, remote control and the like.
The structure of the system power circuit is shown in fig. 5. The 12V power supply is used as an input power supply, and shunt is performed. One path is used for controlling the electromagnetic valve, and the on-off logic of the electromagnetic valve is realized by adopting a relay; after the other path is stabilized, the voltage is converted into 5V for supplying power to the wireless network communication module, the sensor module and the relay module, and further converted into 3.3V for supplying power to other peripheral equipment in the main control chip.
The above-mentioned embodiments only express several embodiments of the present invention, and the description thereof is specific and detailed, but not to be understood as limiting the scope of the present invention. It should be noted that various changes and modifications can be made by those skilled in the art without departing from the spirit of the invention, and these changes and modifications are all within the scope of the invention. Therefore, the protection scope of the present patent should be subject to the appended claims.
Claims (10)
1. A farmland irrigation intelligent decision system based on digital twinning is characterized by comprising a physical layer, a data acquisition layer, a twinning model layer, a functional layer and an application layer;
the physical layer comprises sprinkling irrigation equipment and hardware equipment, wherein the hardware equipment comprises a soil temperature and humidity sensor, a camera, an on-site meteorological station, an information acquisition gateway and a controller;
the data acquisition layer is used for acquiring farmland information and equipment running states and comprises a data acquisition module and a remote communication module;
the twin model layer is used for twin data fusion and construction of a digital twin model;
the functional layer comprises a visual real-time monitoring platform, a precise irrigation decision-making system and an intelligent irrigation control system, the precise irrigation decision-making system makes an intelligent decision on the irrigation quantity and the irrigation time based on air temperature and humidity, wind speed, sunshine hours and pipeline flow data and by means of an irrigation decision-making algorithm, and the intelligent irrigation control system forms an instruction for remotely controlling irrigation equipment to perform irrigation action according to the actual required irrigation time obtained by the irrigation decision-making system;
the system application layer develops an interface on a UI interface, and realizes three-dimensional display, equipment maintenance, irrigation decision, intelligent early warning, real-time monitoring, chart statistics and remote control.
2. The intelligent decision-making system for farmland irrigation based on the digital twin as claimed in claim 1, wherein the data acquisition module is used for acquiring data of field sensors with different protocols and supporting access of analog quantity, digital quantity and RS485 data; the data acquisition module and the remote communication module are communicated through a Lora wireless network, the remote communication module packs data of each data acquisition device, then processes the data into MQTT protocol data, and data transmission is realized through a 4G/5G, ethernet or WiFi and a cloud platform.
3. The intelligent digital twin-based farmland irrigation decision making system as claimed in claim 1, wherein said twin data fusion is performed on a time scale and a space scale, and on the time scale, negligence errors of the data are eliminated by using a distribution diagram method aiming at temperature, humidity and illumination intensity data parameters; on the spatial scale, the temperature and humidity, the pipeline pressure and the flow data are fused by a weighted fusion algorithm to obtain accurate parameters, and the fused data are input into the digital twin model.
4. The intelligent decision making system for farmland irrigation based on digital twins as claimed in claim 3, wherein the said distribution map method comprises the following steps:
s1: sequencing n measurement results measured by m sensors in a farmland from small to large to obtain a measurement sequence: x = (X) 1 ，x 2 ，…，x n )
S2: when n is an even number, a median value x is defined N Is composed of
When n is odd, define median x N Comprises the following steps:
s3: upper quartile F On the upper part Is [ x ] N ，x n ]Median, lower quartile F in interval Lower part Is [ x ] 1 ，x N ]The dispersion of the median and quartile in the interval is dF = F On the upper part -F Lower part ；
S4, setting the effective judgment interval to be F Lower part –dF，F Upper part of +dF]If the data is in the interval, the data is judged to be valid, and if the data is out of the interval, the data is judged to be negligent error and is rejected.
5. The intelligent decision making system for digital twin based field irrigation according to claim 1, wherein said building a digital twin model comprises the steps of:
step T1: oblique photography pictures of the farmland by using an unmanned aerial vehicle, and combining global digital elevation data to synthesize and process the pictures by using a Dajiang intelligent map software tool, so as to realize the construction of a three-dimensional scene of the farmland;
and step T2: establishing three-dimensional models of a plurality of physical entities in an intelligent farmland scene by using a three-dimensional modeling tool, and solving structural parameters, geometric parameters, material parameters, state parameters and boundary conditions of the three-dimensional models by using a finite element analysis method;
step T3: rendering and adding materials to the model by using a 3D Max three-dimensional rendering tool according to the three-dimensional model obtained in the step T2, and optimizing the edge part of the model;
and step T4: and (3) introducing the model rendered in the step T3 into a three-dimensional scene through three-dimensional visualization management software TerraExplorer, giving physical, geometric, behavioral and regular dynamic attributes to the digital twin model, and dynamically reflecting various running states of the entity irrigation equipment.
6. The intelligent decision-making system for farmland irrigation based on digital twins as claimed in claim 1, wherein the visual real-time monitoring platform is used for acquiring environmental parameter information and soil temperature, humidity and soil moisture information acquired by a weather station and transmitting the acquired information to the precision irrigation decision-making system.
7. The intelligent farmland irrigation decision-making system based on the digital twinning as claimed in claim 1, wherein the intelligent irrigation decision-making system obtains the irrigation quantity required by crops through an irrigation decision-making algorithm according to the collected data information of the weather station, the temperature and humidity of the crop soil, the wind speed, the sunshine hours and the pipeline flow, and then obtains the actual required irrigation time by combining the pipeline flow to form an irrigation decision-making control strategy.
8. The intelligent digital twin-based field irrigation decision system as claimed in claim 7 wherein said irrigation decision algorithm comprises:
step A1: firstly, calculating the reference crop water demand ET according to a Penman-Monteith correction formula (1) 0 ；
Wherein, ET 0 : reference crop evapotranspiration amount, mm/d;
t: calculating the average air temperature 2m away from the ground within a time period;
Δ: the slope of the plant rising pressure curve, kPa/DEG C;
R n : net solar radiation, MJ/(m) 2 ·d)；
G: soil heat flux, MJ/(m) 2 ·d)；
γ: monitored wet and dry constant, kPa/° C;
e s : saturated water pressure, kPa; e.g. of the type a : actual water gas pressure, kPa;
u 2 : average wind speed at 2m height from the ground, m/s;
step A2: calculating to obtain the actual water demand Etc of the crops by combining the water storage coefficient Kc of the growth stage of the crops, wherein the Kc refers to the FAO recommended value and is set according to the growth stage of the crops;
ET c ＝K c ET 0 (2)
a3: calculating the irrigation quantity per unit area according to the air temperature and humidity, the air speed and the sunshine hours at a position 2m away from the ground, calculating the total irrigation quantity according to the irrigation quantity per unit area and the land area, and finally calculating the irrigation duration according to the collected pipeline flow;
and S4, sending the irrigation amount and the irrigation time to a system control module through a remote place.
9. The intelligent decision making system for digital twin-based field irrigation according to claim 1, wherein the irrigation control strategy employs a soil moisture prediction model to predict soil moisture at the next time point, determines an irrigation strategy by comparing the soil moisture at the next time point with an optimum growth interval, performs dimensionality reduction on the collected data set using PCA, and separately builds RBF prediction models of soil moisture using the data sets before and after dimensionality reduction; calling a newrbe function to design a radial basis network, and setting the expansion speed spread to be 30.
10. The intelligent decision making system for farmland irrigation based on digital twinning as claimed in claim 9, wherein the method for establishing the RBF soil water content prediction model is as follows:
B1：determining a weighting factor W1 from the input layer to the radial base layer, the weighting factor W1= R T ；
B2: solving for an offset b1 from the input layer to the radial base layer by the formula b1= sqrt (-log (.5))/spread;
b3: calculating The distance between each sample and The radial basis neuron according to The formula of The dist function, namely | | | x-W1| |;
b4: calculating the parameter n by a formula n = | | | x-W1| | | | | b1;
b5: repeating the step B1 to the step B4, and then carrying out a step B6;
b6: by the formula Transfer function:
calculating an output result A {1} of the first layer;
b7: (ii) linear expressions [ W {2,1} b, face 2} ] [ A {1}; ones (1, q) ] = T solves for the weight coefficient W2 and the offset b2 of the second layer;
s8: finishing;
wherein, the input is R = [ T = c ,T air ,RH,R n ,V]The output is T = [ theta ]]
Where R is the input vector, T is the output vector, Q is the number of input vectors, a {1} is the output result of the first layer, b1 is the deviation of the input layer from the radial base layer, W1 is the weight of the input layer from the radial base layer, b2 is the deviation of the radial base layer from the output layer, and W2 is the weight of the radial base layer from the output layer.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211508391.9A CN115804334A (en) | 2022-11-28 | 2022-11-28 | Farmland irrigation intelligent decision system based on digital twins |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211508391.9A CN115804334A (en) | 2022-11-28 | 2022-11-28 | Farmland irrigation intelligent decision system based on digital twins |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115804334A true CN115804334A (en) | 2023-03-17 |
Family
ID=85484364
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211508391.9A Pending CN115804334A (en) | 2022-11-28 | 2022-11-28 | Farmland irrigation intelligent decision system based on digital twins |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115804334A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116267543A (en) * | 2023-05-05 | 2023-06-23 | 安徽农业大学 | Intelligent monitoring system for identifying soil moisture content of high-flux rice drought resistance |
| CN116738766A (en) * | 2023-08-11 | 2023-09-12 | 安徽金海迪尔信息技术有限责任公司 | Intelligent agriculture online industrialization service system based on digital twinning |
| CN116889141A (en) * | 2023-05-31 | 2023-10-17 | 上海华维可控农业科技集团股份有限公司 | Digital high-standard farmland management system based on digital twin technology |
| CN116975789A (en) * | 2023-09-21 | 2023-10-31 | 北京建工环境修复股份有限公司 | Intelligent farmland field analysis method, system and medium based on big data |
| CN117272247A (en) * | 2023-11-17 | 2023-12-22 | 沧州师范学院 | Data integration method and system applied to digital twin intelligent village |
-
2022
- 2022-11-28 CN CN202211508391.9A patent/CN115804334A/en active Pending
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116267543A (en) * | 2023-05-05 | 2023-06-23 | 安徽农业大学 | Intelligent monitoring system for identifying soil moisture content of high-flux rice drought resistance |
| CN116889141A (en) * | 2023-05-31 | 2023-10-17 | 上海华维可控农业科技集团股份有限公司 | Digital high-standard farmland management system based on digital twin technology |
| CN116738766A (en) * | 2023-08-11 | 2023-09-12 | 安徽金海迪尔信息技术有限责任公司 | Intelligent agriculture online industrialization service system based on digital twinning |
| CN116738766B (en) * | 2023-08-11 | 2023-10-13 | 安徽金海迪尔信息技术有限责任公司 | Intelligent agriculture online industrialization service system based on digital twinning |
| CN116975789A (en) * | 2023-09-21 | 2023-10-31 | 北京建工环境修复股份有限公司 | Intelligent farmland field analysis method, system and medium based on big data |
| CN116975789B (en) * | 2023-09-21 | 2023-12-05 | 北京建工环境修复股份有限公司 | Intelligent farmland field analysis method, system and medium based on big data |
| CN117272247A (en) * | 2023-11-17 | 2023-12-22 | 沧州师范学院 | Data integration method and system applied to digital twin intelligent village |
| CN117272247B (en) * | 2023-11-17 | 2024-02-02 | 沧州师范学院 | Data integration method and system applied to digital twin intelligent village |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115804334A (en) | Farmland irrigation intelligent decision system based on digital twins | |
| CN104904569B (en) | A kind of intelligent irrigation regulator control system and method based on the estimation of dynamic water content | |
| CN107945042B (en) | Crop growth irrigation decision control system | |
| CN206833217U (en) | A kind of field planting monitoring system | |
| CN106688827B (en) | A kind of irrigation decision system and method based on agricultural system model | |
| CN105830870B (en) | A kind of long distance wireless farmland monitoring system and method | |
| CN110050673A (en) | A kind of intelligent irrigation management system | |
| CN104731135A (en) | Control device and method for family farm | |
| CN110134147A (en) | A kind of autonomous paths planning method and device of plant protection drone | |
| CN109452146A (en) | Water-saving Irrigation of Winter Wheat decision-making technique, control device and control system | |
| CN113841595A (en) | Energy-saving intelligent accurate irrigation system based on combination of soil humidity and depth | |
| CN106383536A (en) | Greenhouse smart control system and method | |
| CN113919615A (en) | Farmland intelligent irrigation decision-making system based on unmanned aerial vehicle remote sensing data inversion | |
| CN112136667B (en) | Intelligent sprinkling irrigation method and system based on edge machine learning | |
| Suciu et al. | Efficient IoT system for precision agriculture | |
| WO2023173903A1 (en) | Remote intelligent variable-rate irrigation apparatus for photovoltaic-driven translational sprinkler | |
| CN115362811A (en) | Mountain crop intelligent cultivation system based on digital twinning | |
| CN116360331A (en) | Universal automatic control system and control method | |
| CN113919964A (en) | Intelligent agricultural greenhouse management system and method based on BIM | |
| CN113110036A (en) | Agricultural wireless automatic monitoring control system based on FOD-PID control method | |
| CN109213240A (en) | A kind of strawberry greenhouse wireless monitor and control system based on self adaptive control | |
| CN206380441U (en) | A kind of irrigation rig | |
| CN110579961B (en) | Three-dimensional planting-oriented garden intelligent water supply method and system | |
| CN115633622A (en) | Intelligent orchard irrigation system and method | |
| CN206115670U (en) | System for automated analysis crop output influence factor |
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 |