Variable Rate Liquid Fertilizer Applicator for Deep-fertilization in Precision Farming Based on ZigBee Technology

Variable Rate Liquid Fertilizer Applicator for Deep-fertilization in Precision Farming Based on ZigBee Technology

6th IFAC Conference on Sensing, Control and Automation for Available online at www.sciencedirect.com 6th IFAC Conference on Sensing, Control and Autom...

732KB Sizes 0 Downloads 10 Views

6th IFAC Conference on Sensing, Control and Automation for Available online at www.sciencedirect.com 6th IFAC Conference on Sensing, Control and Automation for Agriculture 6th IFAC on Control 6th IFAC Conference Conference on Sensing, Sensing, Control and and Automation Automation for for Agriculture December 4-6, 2019. Sydney, Australia Agriculture Agriculture4-6, 2019. Sydney, Australia December December December 4-6, 4-6, 2019. 2019. Sydney, Sydney, Australia Australia

ScienceDirect

IFAC PapersOnLine 52-30 (2019) 43–50

Variable Rate Liquid Fertilizer Applicator for in Precision Variable Rate Liquid Fertilizer Applicator for for Deep-fertilization Deep-fertilization in Precision Precision Variable Rate Liquid Fertilizer Applicator Deep-fertilization in Variable Rate Liquid Fertilizer Applicator forTechnology Deep-fertilization in Precision Farming Based on ZigBee Farming Based Based on ZigBee ZigBee Technology Farming Farming Based on on ZigBee Technology Technology

Xue Xiuyun, Xu Xufeng, Zhang Zelong, Zhang Bin, Song Shuran*, Li Zhen, Hong Tiansheng, Huang Huixian Xue Xiuyun, Xu Xufeng, Zhang Zelong, Zhang Bin, Song Shuran*, Li Zhen, Hong Tiansheng, Huang Huixian Xue Xue Xiuyun, Xiuyun, Xu Xu Xufeng, Xufeng, Zhang Zhang Zelong, Zelong, Zhang Zhang Bin, Bin, Song Song Shuran*, Shuran*, Li Li Zhen, Zhen, Hong Hong Tiansheng, Tiansheng, Huang Huang Huixian Huixian *College of Electronic Engineering, South China Agricultural University, Guangzhou, 510642,China *College of Electronic Engineering, South China Agricultural University, Guangzhou, 510642,China (Corresponding author, Tel: +86China 20 85282269, e-mail:[email protected]) *College Electronic South Agricultural University, *College of of Electronic Engineering, Engineering, South Agricultural University, Guangzhou, Guangzhou, 510642,China 510642,China (Corresponding author, Tel: +86China 20 85282269, e-mail:[email protected]) (Corresponding author, Tel: +86 20 85282269, e-mail:[email protected]) (Corresponding author, Tel: +86 20 85282269, e-mail:[email protected]) Abstract: In this article, a variable rate liquid fertilizer applicator for deep-fertilization based on ZigBee Abstract: In this article, a variable rate liquid fertilizer applicator for deep-fertilization based on ZigBee technology A hostrate computer with remote control and a STM32F103RET6 Abstract: Inwas this article, liquid applicator for deep-fertilization based Abstract: thiselaborated. article, aa variable variable liquid fertilizer fertilizer applicator for software deep-fertilization based on on ZigBee ZigBee technologyInwas elaborated. A hostrate computer with remote control software and a STM32F103RET6 microcontroller were combined to measure and control the liquid fertilizer output: monitoring the liquid technology was elaborated. A host computer with remote control software and aa STM32F103RET6 technology was elaborated. A host computer with remote control software and STM32F103RET6 microcontroller were combined to measure and control the liquid fertilizer output: monitoring the liquid fertilizer level and flow thefertilizer flow meter, andmonitoring an incremental PID microcontroller were combined to measure and control liquid output: the microcontroller werecollection combinedthe to liquid measure andinformation control the the by liquid output: the liquid liquid fertilizer level and collection the liquid flow information by thefertilizer flow meter, andmonitoring an incremental PID control algorithm was used to dynamically adjust the converter frequency for reaching the set liquid fertilizer level and collection the liquid flow information by the flow meter, and an incremental PID fertilizer level andwas collection liquid flowadjust information by the frequency flow meter, an incremental PID control algorithm used tothe dynamically the converter forand reaching the set liquid fertilizer flow accurately. Through the field test, the influence of parameters, such as fertilization depth, control algorithm was used to dynamically adjust the converter frequency for reaching the set liquid control algorithm was used to dynamically adjust the converter frequency for reaching the set liquid fertilizer flow accurately. Through the field test, the influence of parameters, such as fertilization depth, pressure of fertilizer injection, and the of valve opening such for water return of depth, liquid frequency of inverter, fertilizer accurately. Through the the parameters, as fertilizer flow flow accurately. Through the field field test, test, the influence influence parameters, as fertilization fertilization frequency of inverter, pressure of fertilizer injection, and the of valve opening such for water return of depth, liquid fertilizer pump, on precise flow control is analyzed, and the mathematical model is established by using frequency of inverter, pressure of fertilizer injection, and the valve opening for water return of liquid frequencypump, of inverter, pressure fertilizer injection, and valve opening for iswater return of fertilizer on precise flow of control is analyzed, and thethe mathematical model established by liquid using experimental data. The results showed that, the applicator accuracy is able to reach 99.52% and the liquid fertilizer pump, on precise flow control is analyzed, and the mathematical model is established by using fertilizer pump, on The precise flow controlthat, is analyzed, and the mathematical model is established by using experimental data. results showed the applicator accuracy is able to reach 99.52% and the liquid -1 fertilizer consumption in eachshowed fertilization process is within 0.2L.min with the99.52% fertilizer application experimental data. The results that, the applicator accuracy is able reach and the liquid -1, to experimental data. The results showed that, the applicator accuracy is able to reach 99.52% and the liquid fertilizer consumption in each fertilization process is within 0.2L.min -1, with the fertilizer application -1 depth changing, the maximum flow output process difference is within 0.2L.min inverterapplication frequency -1, with fertilizer consumption in is 0.2L.min the fertilizer -1 and fertilizer consumption in each each fertilization fertilization is within within 0.2L.min , with thethe depth changing, the maximum flow output process difference is within 0.2L.min thefertilizer inverterapplication frequency -1 and difference 1Hz. When the valve opening for back water is 40%, the system works most stable is within -1 depth changing, the maximum flow output difference is within 0.2L.min and the inverter frequency depth changing, the1Hz. maximum outputopening difference is within the inverter frequency difference is within Whenflow the valve for back water0.2L.min is 40%, theand system works most stable and the flow output error is the smallest. difference is within 1Hz. When the valve opening for back water is 40%, the system works most difference is output within error 1Hz. isWhen the valve opening for back water is 40%, the system works most stable stable and the flow the smallest. and the flow output error is the smallest. and the flow output error is the smallest. © 2019, IFACPrecision (International Federation of Automatic Hosting by Elsevier Ltd. All rightsNetwork, reserved. Liquid Keywords: Agriculture and VariableControl) Rate Technologies, Wireless Sensor Keywords: Precision Agriculture and Variable Rate Technologies, Wireless Sensor Network, Liquid fertilizer, Deep-fertilization, ZigBee Keywords: Precision Agriculture and Variable Rate Technologies, Wireless Sensor Network, Liquid Keywords:Deep-fertilization, Precision Agriculture fertilizer, ZigBeeand Variable Rate Technologies, Wireless Sensor Network, Liquid fertilizer, Deep-fertilization, ZigBee fertilizer, Deep-fertilization, ZigBee Liang Chunying et al. (2010) discussed the effects of 1. INTRODUCTION Liang Chunying et al. (2010) discussed the effects of parameters such aset liquid fertilizer concentration and pipe 1. INTRODUCTION Liang Chunying al. discussed the of Liang Chunying al. (2010) (2010) discussed the effects effects of parameters such aset liquid fertilizer concentration and pipe 1. INTRODUCTION 1. INTRODUCTION pressure on the performance of precision liquid fertilizer parameters such as liquid fertilizer concentration and pipe Liquid fertilizer is different from simple water and fertilizer parameters such as liquid fertilizer concentration and pipe pressure on the performance of precision liquid fertilizer Liquid fertilizer different from in simple water and fertilization Most of them were experimented in the the of liquid mixture. It can beis applied the efficient andfertilizer precise pressure Liquid is widely different from simple water fertilizer pressure on on system. the performance performance of precision precision liquid fertilizer fertilizer system. Most of them were experimented in the Liquid fertilizer fertilizer different from in simple water and and mixture. It can beis widely applied the efficient andfertilizer precise fertilization form of liquid fertilizer surface spray, but the research on fertilization system. Most of them were experimented in the fertilization with the prescription chart according to the needs mixture. It can be widely applied in the efficient and precise fertilization system. Most of them were experimented inroot the form of liquid fertilizer surface spray, but the research on root mixture. It can be widely applied in the efficient and precise fertilization with the prescription chart according to the needs form application was insufficient. At the same time, the traditional of liquid fertilizer surface spray, but the research on root of crops (Mori et al. 2010). The application of liquid organic fertilization with the prescription chart according to the needs form of liquid fertilizer surface spray, but the research on root application was insufficient. At the same time, the traditional fertilization with the prescription chart according to the needs of crops (Mori et al. 2010). The application of liquid organic agricultural information measurement control systems insufficient. At time, the fertilizer can improve soil organic matter, soil structure and application of crops et The of organic application was was insufficient. At the the same sameand time, the traditional traditional information measurement and control systems of crops (Mori (Mori et al. al. 2010). 2010). The application application of liquid liquid organic fertilizer can improve soil organic matter, soil structure and agricultural have the disadvantages of complicated wiring, high agricultural information information measurement measurement and and control control systems systems the capacity of holding water (Chiang etsoil al. structure 2016). Deep fertilizer can improve soil organic matter, and agricultural have the disadvantages of complicated wiring, high fertilizer can improve soil organic matter, soil structure and the capacity of holding water (Chiang et al. 2016). Deep construction and maintenance costs, and inconvenient power the of complicated wiring, high application fertilizer a fertilization method in have the ofliquid holding water (Chiang et Deep the disadvantages disadvantages ofcosts, complicated wiring, power high construction and maintenance and inconvenient the capacity capacityof holding water is (Chiang et al. al. 2016). 2016). Deep application ofofliquid fertilizer is a fertilization method in have supply, which ismaintenance not conducive to and real-time and efficient construction and costs, inconvenient power which liquid fertilizer is directly applied to the soil to supply application of liquid fertilizer is a fertilization method in construction and maintenance costs, and inconvenient power which is not conducive to real-time and efficient application liquid is fertilizer a fertilization in supply, which liquidoffertilizer directlyisapplied to the soilmethod to supply acquisition of is monitoring data to and preciseand control of which not real-time efficient roots plants. Compared with soil surface deep supply, which liquid fertilizer is applied to the which not conducive conducive real-time efficient acquisition of ismonitoring data toand preciseand control of whichof fertilizer is directly directly applied to fertilization, the soil soil to to supply supply roots ofliquid plants. Compared with soil surface fertilization, deep supply, agricultural systems in complex terrain and large-scale acquisition of monitoring data and precise control of application of liquid fertilizer can improve the absorption of roots of Compared with soil surface deep acquisition of monitoring data and precise of systems in complex terrain and control large-scale roots of plants. plants. Compared withcan soilimprove surface fertilization, fertilization, deep application of liquid fertilizer the absorption of agricultural operating areas. Therefore, ZigBee wireless communication agricultural systems in complex terrain and large-scale fertilizer by plant roots better, prevent roots from floating, application of liquid fertilizer can improve the absorption of agricultural systems in complex terrain and large-scale operating areas. Therefore, ZigBee wireless communication application of liquid fertilizer can improve the absorption of fertilizer by plant roots better, prevent roots from floating, operating technologyareas. is widely used in the design of communication wireless sensor Therefore, ZigBee wireless enhance the resistance crops, roots and reduce the loss technology fertilizer by plant better, prevent from areas. Therefore, ZigBee wireless fertilizer by lodging plant roots roots better, of prevent from floating, floating, is intelligent widely used in the design of communication wireless sensor enhance the lodging resistance of crops, roots and reduce the loss operating networks agricultural irrigation systems and technologyinis is widely widely used used in the the design design of wireless wireless sensor of fertilizer and environmental pollution caused by the enhance the lodging resistance of crops, and reduce the loss technology in of sensor enhance the lodging resistance of crops, and reduce the loss networks in intelligent agricultural irrigation systems and of fertilizer and environmental pollution caused by the agricultural information measurement and control systems, networks in in intelligent intelligent agricultural agricultural irrigation irrigation systems systems and and volatilization fertilizer (CHEN et al.caused 2013, da of fertilizer and environmental pollution by the networks of fertilizer of andliquid environmental pollution bySilva the agricultural information and control systems, volatilization of liquid fertilizer (CHEN et al.caused 2013, da Silva which has the advantagesmeasurement of low equipment cost, timely and agricultural information measurement and control systems, et al. 2017). volatilization of liquid fertilizer (CHEN et al. 2013, da Silva agricultural information measurement and control systems, volatilization has thetransmission advantages (Choudhury of low equipment cost, Sheng timely etand et al. 2017). of liquid fertilizer (CHEN et al. 2013, da Silva which reliable data et al. 2015, al. which the advantages of cost, timely et al. 2017). which has has thetransmission advantages (Choudhury of low low equipment equipment cost, Sheng timely etand and et al. 2017). reliable data et al. 2015, al. Wang Jinwu et al. (Wang et al. 2015, Lang et al. 2013, Feng reliable 2012, Zhao et al. 2016). data transmission (Choudhury et al. 2015, Sheng et al. reliableZhao dataettransmission Wang Jinwu et al. (Wang et al. 2015, Lang et al. 2013, Feng 2012, al. 2016). (Choudhury et al. 2015, Sheng et al. of liquid et al. 2017) are dedicated to the design and testing Wang Jinwu et al. (Wang et al. 2015, Lang et al. 2013, Feng 2012, Zhao et al. 2016). Wang Jinwuare et al. (Wang et Lang al. 2013, Feng 2012, Zhao aet variable al. 2016).liquid fertilizer applicator for deepet al. 2017) dedicated to al. the2015, design andettesting of liquid fertilizer deep theof et al. al. 2017) 2017) are application dedicated to tomachines, the design designespecially and testing testing of design liquid Therefore, Therefore, a variable liquid fertilizer applicator for deepet are dedicated the and liquid fertilizer deep application machines, especially the design fertilization on ZigBee has been developed. variable liquid fertilizer for and optimization of key machines, components of thethe fertilizer deep application application machines, especially theacupoint design Therefore, Therefore, aabased variable liquid technology fertilizer applicator applicator for deepdeepfertilization based on ZigBee technology has been developed. fertilizer deep especially design and optimization of key components of the acupoint fertilization based on ZigBee technology has been developed. fertilization mechanism. Wang et al. (2017) have developed a and optimization optimization of of key key components components of of the the acupoint acupoint fertilization based on ZigBee technology has been developed. and fertilization mechanism. al. (2017) 2. METHODS AND MATERIALS backpack liquid fertilizer,Wang whichet drill thehave holedeveloped and adjustaaa fertilization mechanism. Wang et can al. have developed fertilization mechanism. al. (2017) (2017) 2. METHODS AND MATERIALS backpack liquid fertilizer,Wang whichet can drill thehave holedeveloped and adjust 2. the amount of fertilization. Due to the high requirements of backpack liquid fertilizer, which can drill the hole and adjust 2. METHODS METHODS AND AND MATERIALS MATERIALS backpack liquid fertilizer, which drill the requirements hole and adjust the amount of fertilization. Due can to the high of agricultural machine and agronomy, the current research on the amount of fertilization. Due to the high requirements of 2.1 Hardware Architecture of the System the amount of fertilization. Due to the requirements of agricultural machine and agronomy, thehigh current research on 2.1 Hardware Architecture of the System liquid fertilizer variable fertilizationthe technology is mainly agricultural machine and current on Hardware agricultural machine and agronomy, agronomy, current research research on 2.1 liquid fertilizer variable fertilizationthe technology is mainly Hardware Architecture Architecture of of the the System System applied to the fertilization of crop leaves and stems.is Limainly et al. 2.1 liquid variable technology The liquid fertilizer variable deep application experiment liquid fertilizer fertilizer variable fertilization fertilization technology applied to the fertilization of crop leaves and stems.is Limainly et al. The liquidis fertilizer variable deep (2018) the adaptive fuzzy PID liquid fertilizer appliedhave to the thepresented fertilization of crop crop leaves leaves and stems. stems. Li et et al. al. platform mainly composed by aapplication frequency experiment conversion applied to fertilization of and Li The variable application experiment (2018) have presented the adaptive fuzzy PID liquid fertilizer The liquid liquidis fertilizer fertilizer variable deep deep application experiment platform mainly composed by a frequency conversion fertilization technology based on the traditional proportional (2018) have presented the adaptive fuzzy PID liquid fertilizer control system, a communication system, a monitoring (2018) have presented the adaptive fuzzy PID liquid fertilizer platform is mainly composed by a frequency conversion platform system, is mainly composed by a system, frequency conversion fertilization technology the traditional proportional a fertilizer communication a monitoring algorithm. Zhang et al. based (2015)on designed a precise control of control fertilization technology based on the traditional proportional system and a liquid system. fertilization technology based on the traditional proportional control system, a communication system, a monitoring algorithm. Zhang et al. (2015) designed a precise control of system control system, a communication system, a monitoring and a liquid fertilizer system. liquid fertilizer output system usingaa aprecise microcontroller. algorithm. Zhang et (2015) designed control algorithm. Zhangflow et al. al. (2015) designed control of of system liquid fertilizer flow output system using aprecise microcontroller. system and and aa liquid liquid fertilizer fertilizer system. system. liquid fertilizer flow output system using a microcontroller. liquid fertilizer flow output system using a microcontroller.

2405-8963 © 2019, IFAC (International Federation of Automatic Control) Hosting by Elsevier Ltd. All rights reserved. Peer review©under of International Federation of Automatic Copyright 2019 responsibility IFAC 43 Control. 10.1016/j.ifacol.2019.12.487 Copyright © 2019 IFAC 43 Copyright 43 Copyright © © 2019 2019 IFAC IFAC 43

IFAC AGRICONTROL 2019 44 December 4-6, 2019. Sydney, Australia

Xue Xiuyun et al. / IFAC PapersOnLine 52-30 (2019) 43–50

Fig. 1 is the diagram of the overall hardware of the system. The control signal sent by the PC is transmitted to the microcontroller through the ZigBee. After receiving the signal, the microcontroller sends the corresponding control instruction to the inverter. The inverter resolves and responds to the received instruction and begins to adjust the motor speed of the liquid fertilizer pump to change the output of the liquid fertilizer system. The flow meter and pressure transmitter collect corresponding data and feedback to the microcontroller. After the PID operating, the microcontroller performs the next control action according to the operation result. During the operation of the system, the liquid level transmitter detects the remaining amount of liquid fertilizer. When the remaining amount is lower than the set value, an alarm is triggered and the electromagnetic valve connected to the output pipe of the fertilizer storage tank is closed. PC control signal

Beep alarm and close the solenoid valve

Transmited by ZigBee

MCU send the signal

PID calculation

Inverter response the signal

The system calculates the current flow value based on the sampled real-time voltage value. The pressure transmitter works in the same way as the liquid level transmitter. 24V power supply and 4 to 20mA current represent the mapping range of the corresponding pressure value and liquid level. The system can get the current pressure value and liquid level height by collecting the current change and calculating the current. 2.1.4 Liquid Fertilizer System Fig. 2 is the structure of the liquid fertilizer gun used in the system. It is made of high-strength stainless steel and can withstand large pressures of fertilizer. The fertilizer gun has a depth limit, which makes the fertilizer depth adjustment convenient and suitable for fertilization operations under different soil depths. The fertilizer liquid is injected into the soil by high pressure, so the contact area with the root of the crop is large, improving the utilization rate of the fertilizer and accelerating the absorption (Wang et al. 2005).

Level transmitter monitor the rest

Adjust the spray pump motor speed

Liquid fertilizer system perform actions

Flowmeter and pressure transmitter collect the data

Fig. 1. System structure of variable liquid fertilizer applicator for deep-fertilization. 2.1.1 Frequency Conversion Control System

Fig. 2. Liquid manure gun structure diagram

The liquid fertilizer variable system uses STM32F103RET6 as the control terminal. uses the Delta Inverter VFD-CP2000 to control motor the microcontroller sends the control instruction to the inverter through the wireless communication device, and the inverter performs frequency conversion control on the spray pump motor after receiving the corresponding instruction.

2.2 Software Architecture of the System 2.2.1 Design of the PC Software In order to realize variable fertilization, a variable deep control software was designed on the PC. The software is written in C/C++ language in the QT compiler environment. It combines Microsoft Excel and QwtPlot controls to add the function of data exporting and real-time displaying of dynamic data curves. During the working process, the system sends control information to the microcontroller through the PC to indirectly control the variable fertilization mechanism.

2.1.2 Communication System The experiment platform is equipped with a PC host to assist control, and a ZigBee communication module is used as a data network relay station for both the host computer and the microcontroller. The ZigBee networking uses a star topology, the co-control PC acts as a coordinator node, and multiple microcontrollers act as terminal nodes.

2.2.2 Design of the PLC Software The stable operation of the system requires the cooperation of all parts of the hardware, so hardware programming is performed on the STM32F103RET6. The software is written in the Keil μVision5 compilation environment, mixed with C language and partial assembly language. These include hardware drivers, analog-to-digital conversion configurations, communication packet encapsulation, and user interaction.

2.1.3 Monitoring System It is necessary to monitor the liquid fertilizer reserve and flow rate when the liquid fertilizer system is running. In addition, the liquid fertilizer output pressure should not exceed the preset threshold. Therefore, the experiment platform is equipped with a liquid level transmitter to monitor the remaining amount of the fertilizer tank, and when the remaining amount is lower than the set value, an alarm is triggered and the electromagnetic valve is closed. At the same time, the flow meter and pressure transmitter connected to the spray pump outlet feedback the changing voltage analog to the microcontroller. According to the Analog-toDigital Converter and the flow pressure formula, the microcontroller can obtain the real-time flow and pressure values.

2.3 Key Technology of the System 2.3.1 ZigBee Ad Hoc Network Technology Fig. 3 shows the star network topology of ZigBee. It consists of a Co-ordinator and several End Devices. The Coordinator functions as a sink node in the network. It is a network coordinator and can communicate with terminal nodes in the network.

44

IFAC AGRICONTROL 2019 December 4-6, 2019. Sydney, Australia

Xue Xiuyun et al. / IFAC PapersOnLine 52-30 (2019) 43–50

End Device

45

2.4 Design of the Experiment Plan

End Device

End Device

2.4.1 Conditions and Parameters of the Experiment

Co-ordinator End Device

In order to analyse the various operating performance of the system, the system real-time liquid fertilizer flow value, system pressure value, inverter frequency value and liquid fertilizer consumption are measured and tested. By collecting these key datas, the effects of fertilization depth and system pressure on the flow rate of the liquid fertilizer and the frequency of the inverter are analysed. At the same time, the accuracy of the fertilization amount (the fluctuation range of the real-time value of the liquid fertilizer flow and the preset value) and the fertilization amount deviation were recorded during the experiment, which was the evaluation index of the variable fertilization system. It is necessary to ensure that the system is operating in the safety environment before conducting the experiment.

End Device End Device

End Device

Fig. 3. Star topology diagram of ZigBee Fig. 4 shows the wireless sensor network of the liquid fertilizer variable control system. After the coordinator node is powered on, the network initialization operation is performed, including the ZigBee protocol stack and the hardware peripherals. Then the channel query is performed and a suitable channel is selected to establish a wireless standard network. PAN ID is set to wait for the joining of the End Devices. After End Devices joins the network, the data frame sent by the End Devices is accepted, displayed in real time and stored in the memory. Finally, the data frame is transmitted to the PC through the serial port to complete the display of the data on the PC. Master node(Coordinator)

Child node(End Device)

ZigBee module

MCU Sensors collect the analog data

Data interface A/D conversion

In order to determine the aperture of the back water valve, the system needs to be started from the maximum pressure, that is turn off the fertilizing gun, and set the inverter frequency to 50Hz at the widest aperture of the back water valve. Then gradually adjust the aperture of the back water valve so that the pressure values displayed by the pressure transmitter are 2.5 MPa, 2.0 MPa, 1.5 MPa, 1.0 MPa, and 0.5 MPa. Respectively, the aperture of the back water valve is set to 20%, 40%, 60%, 80%, and 100%, at this time.

serial communication

ZigBee transmission

PC (Display and store the data)

ZigBee module

2.4.2 Effect on the depth of the fertilization The experiment platform applies variable fertilization to the crops and directly transmits the liquid fertilizer to the roots of the crops to increase the absorption of fertilizer through the crop roots. Crops can be divided into shallow roots, medium roots and deep roots according to the different root distribution depths of different crops. In order to simulate crop fertilization scenarios with different root depths, the flow output of the variable fertilization system and the frequency of the inverter were observed under different soil depths.

Fig. 4. Wireless sensor network of variable liquid fertilizer applicator 2.3.2 Control technology of the PID Precise control of flow is a key technology for liquid fertilization (Ge et al. 2016). In order to reach the preset flow value dynamically, the previous flow value of the current position needs to be recorded, so the system uses an incremental PID (Proportion-Integration-Division) algorithm to adjust the flow output.

2.4.3 Effect on pressure of the system The flow rate of the system is controlled by the pressure of the fertilizer, and the pressure of the pump affected by the aperture of the back water valve is closely related to the flow output. Since the variable fertilization system continuously adjusts the liquid fertilizer flow output during operation, that is the real-time fertilizer pressure of the system changes at time to keep the current flow within a preset range. Therefore, the experiment mainly observes the change of the parameters of the system by adjusting the aperture of the back water valve under the premise of preset pressure on the PC.

According to the rule, the PID control flow chart is designed, as shown in Fig.5. Start

Initialize the PID Get presets and errors

Calculate the incremental value

N

Whether the current flow is greater than the preset value

3. RESULTS AND ANALYSIS

Y

N

Current frequency value - absolute increment value > 10Hz? Y

Set minimum frequency value of the inverter

Send the calculated frequency value to the inverter to adjust the frequency

N

Current frequency value + absolute increment value > 50Hz?

3.1 Experiment of the park

Y

Set maximum frequency value of the inverter

3.1.1 Determination of the variable fertilization accuracy During the test, the fertilizer amount needs to be preset in the system of the PC. After clicking the start button and turn on

Fig. 5. Flowchart of PID control 45

IFAC AGRICONTROL 2019 46 December 4-6, 2019. Sydney, Australia

Xue Xiuyun et al. / IFAC PapersOnLine 52-30 (2019) 43–50

the inverter, the liquid fertilizer system starts to work. In the experiment, the fertilization depth is 20cm. The aperture of the back water valve is set to 40%. 50 liquid fertilizer flow set values are selected, and each set value is measured 10 times in order to obtain the average value of the actual fertilization amount. The actual fertilizer amount can be used to calculate the fertilization amount deviation and the accuracy of the system's fertilization. Some experiment data is shown in Table 1.

Table 2. Experiment data of wide back-water valve aperture degree

Table 1. Variable fertilization experiment data and results Preset fertilizatio n amount (Lmin-1) 7.23 9.06 11.21 12.80 15.75 16.70 17.61 18.45

Average actual fertilizatio n amount (Lmin-1) 7.08 8.88 11.07 12.60 15.53 16.62 17.53 18.33

Fertilization amount deviation (Lmin-1)

Aperture of the back water valve (%)

Preset pressure value (MPa)

Average flow value(Lmin-1)

Average frequency value (Hz)

h=10

h=50

h=10

h=50

40

0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0

7.23 9.10 10.92 12.73 14.25 15.76 16.62 17.69 18.32

7.25 9.07 11.02 12.83 14.18 15.88 16.50 17.57 18.42

10.00 12.68 14.89 17.45 19.49 21.29 25.28 36.01 48.84

10.00 12.12 15.66 17.99 19.14 21.55 25.28 36.41 48.53

Accuracy (%)

Tips:h is the depth of fertilization, the unit is cm.

0.15 0.18 0.14 0.20 0.22 0.08 0.08 0.12

97.93 98.01 98.75 98.44 98.60 99.52 99.55 99.35

Table 3. Experiment data of narrow back-water valve aperture degree

The results show that the inverter can change the speeds of the spray pump motor through the control of the variables, which can change the amount of fertilizer applied each time. The average accuracy of the fertilization system can achieve 98.77%. 3.1.2 Determination of parameters under different fertilization depths

Average frequency value (Hz)

Aperture of the back water valve (%)

Preset pressure value (MPa)

h=10

h=50

h=10

h=50

80

0.4 0.5 0.6 0.7 0.8 0.9 1.0

6.83 7.92 8.92 9.97 10.88 11.68 12.52

6.70 8.07 9.00 10.02 11.00 11.80 12.62

10.96 15.34 21.37 28.68 34.71 40.93 47.78

10.95 16.03 21.49 27.89 34.68 40.93 47.61

Average flow value(Lmin-1)

The experiment results show that in the narrow aperture of the back water valve, the average flow rate is larger than the wide back water opening environment, and the average frequency is smaller.

In order to eliminate the influence of system pressure, the experiment was carried out under the premise that the aperture of the back water valve was 40% and 80%. Fan Jun et al. (2013) determined the root depths of shallow roots, medium roots and deep roots plants of about 10 cm, 30 cm and 50 cm according to the water consumption coefficients of different depth roots. Use this as a reference in the experiment, 10 cm under the soil was used as the initial depth and 5 cm was used as the step value, obtaining 9 fertilization depth gradients. The external pressure value was preset by PC software under 9 different gradients, and the deep application experiment was carried out with 0.4 MPa as the initial pressure value and 0.2 Mpa or 0.1 MPa as the step value. The parameters were measured 10 times during the experiment in order to eliminate accidental errors. Some experiment data is shown in Table 2 and Table 3.

3.1.3 Determination of parameters under different system pressures In order to analyze the influence of system pressure on the parameters, the control variable method is used to first preset the pressure of the fertilizer, and in the case where the difference in the back water opening is as large as possible, the position of the back valve opening close to the flow output range is selected. Table 4 shows the output range of the system flow rate under each back water opening. It is more appropriate to select 40% and 80% as the test gradient. After determining the opening position of the back valve, select 0.1MPa or 0.2MPa as the appropriate step distance, divide the corresponding pressure filling pressure interval into several groups, and preset the pressure of the fertilizer in the PC computer software to carry out the system pressure test. In addition, the above tests were carried out at a fertilization depth of 20 cm and each relevant pressure gradient was measured 10 times.

46

IFAC AGRICONTROL 2019 December 4-6, 2019. Sydney, Australia

Xue Xiuyun et al. / IFAC PapersOnLine 52-30 (2019) 43–50

Table 4. Flow output range of each -water valve aperture degree Position of the back-water valve aperture 20% 40% 60% 80% 100%

Table 6. Experiment data difference of 80% back-water valve aperture degree

Flow output range of the system (Lmin-1) 7.47~20.67 7.23~18.45 7.08~15.92 6.73~12.73 1.80~8.08

3.1.4 Determination of liquid fertilizer consumption The experiment was carried out at the aperture of the back water valve of 40%, 80% and 100%, and 6 litres water was used to simulate the application amount of one fertilizer point in actual fertilization. During the experiment, the pressure of the fertilizer is preset in the PC software. And the time taken for each 6 litre of water is calculated according to the flow value of the stable output under the corresponding pressure of the fertilizer. When the fertilizer is finished, the consumption can be calculated by the changing weight of the water.

The experiment data difference values of Tables 5 and 6 can be obtained according to Tables 2 and 3. According to the analysis, the maximum difference of the flow rate of different fertilization depths under the 40% the aperture of the back water valve is 0.12 litres per minute, and the maximum frequency difference is 0.77 Hz. The maximum difference of the flow rate of different fertilization depths under the 80% the aperture of the back water valve is 0.15 litres per minute, and the maximum frequency difference is 0.79Hz. Table 5. Experiment data difference of 40% back-water valve aperture degree Error value of the flow (Lmin-1)

40

0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0

0.02 0.03 0.1 0.1 0.07 0.12 0.12 0.07 0.1

Pressure values (MPa)

Error value of the flow (Lmin-1)

40

0.4 0.5 0.6 0.7 0.8 0.9 1.0

0.13 0.15 0.08 0.05 0.12 0.12 0.1

Error value of the frequency (Hz) 0.01 0.69 0.12 0.79 0.03 0 0.17

Soil macrovoid can characterize the phenomenon of water and solute preferential transport, the water can pass most of the soil matrix through the soil macrovoid and reach the deep soil layer in a short time (Gao et al. 2014). It decreases with the increase of soil depth, and the soil with more soil macrovoid has better connectivity. The roots of southern citrus trees are mainly distributed in the 15 to 55 centimetres below the surface, and their roots are so deep that the upper and lower soil spaces are more likely to form highly connected macrovoid structures (Chen et al. 2015). Although human activities will destroy the soil macrovoid to a certain extent, resulting in changes in soil connectivity, the roots of citrus trees are so deeper that the soil structure of the deep root zone is less disturbed by humans, and the soil of the park is coarse. The infiltration of the irrigating water is very faster (Gao et al. 2014). Therefore, when the experiment is carried out in a depth of 10 to 50 centimetre, the liquid fertilizer transmitting rate is fast, and it is not easy to accumulate and block in the soil, and the effect on the water injection hole of the fertilizer gun is small.

3.2.1 Analysis of relationship between fertilization depth, flow rate and frequency

Pressure values (MPa)

Fertilization depth difference (cm)

With the different depth of fertilization, the underground soil structure and water content (that is the liquid fertilizer concentration) also change accordingly. The result will affect the output of the nozzle of the fertilizer gun, which indirectly affects the flow output of the system and the frequency of the inverter.

3.2 Variables analysis of the system

Fertilization depth difference (cm)

47

So it can be considered that the effect of fertilization depth on the system pressure is minimal, and the variable fertilization system is negligible.

Error value of the frequency (Hz) 0 0.56 0.77 0.54 0.35 0.26 0 0.4 0.31

3.2.2 Analysis of relationship between system pressure, flow and frequency According to the partial test data of Tables 2 and 3, the flow rate and frequency curves under different parameters in Fig. 6 are plotted in MATLAB software.

47

IFAC AGRICONTROL 2019 48 December 4-6, 2019. Sydney, Australia

Xue Xiuyun et al. / IFAC PapersOnLine 52-30 (2019) 43–50

fertilizer increases, which is caused by an increase in the motor load and a decrease in the output power. At the aperture of the back water valve of 40%, the frequency of the inverter is abrupt at a pressure of about 1.6 MPa. After that, the frequency increases and the flow rate increases. This phenomenon is more obvious as the pressure of fertilizer injection increases. It shows that even under the wide aperture of the back water valve, when the pressure of the fertilizer is increased to a certain extent, the increase of the motor load will cause the power output of the motor to decrease sharply and the operation efficiency to decrease. At the aperture of the back water valve of 80%, the frequency of the inverter oscillates within the pressure range of 0.4 to 0.6 MPa. Correspondingly, the flow output in the interval also oscillates. The reason is that the unstable flow and manufacturing problems under small flow cause periodic changes in the motor load, and the motor efficiency is so sensitive to load changes that the motor efficiency fluctuates (Gu et al. 2018). This change is exacerbated by the narrow aperture of the back water valve, resulting in unstable system flow output. The oscillation of system and motor overload will make the liquid fertilizer variable system tend to be unstable. Therefore, the system needs to select the appropriate aperture of the back water valve during normal operation, and the fertilizer pressure should not be too large. 3.2.3 Fertilization amount deviation under different system pressures According to the flow value of the stable output under the pressure of the fertilizer, the time required for the system to output 6 litres of water can be calculated. This time is the most ideal work time, which does not consider the PID adjustment process of the system. If the system can complete the task of filling fertilizer amount to the maximum extent during this period, that is, the target value is 6 litres, indicating that the PID adjustment time is short and the system operation efficiency is high. According to the difference between the current amount of fertilizer and the target value, the response speed of the system can be checked. If the PID adjustment time is too long, the system response will be slow, resulting in the loss of liquid fertilizer.

Fig. 6. Fertilization Pressure-flow curve of different fertilizer application depth The frequency of the inverter is affected by the pressure of the fertilizer and the aperture of the backwater valve. Fig. 6 shows the frequency curve of the different depths corresponding to the pressure of the fertilizer. As the pressure of the fertilizer increase, the frequency of the inverter and the motor speed also increases. In addition, the working range of the system working under different the aperture of the back water valve is 0 to 2 MPa under 40% and 0 to 1MPa under 80%. This indicates that the narrow aperture of the back water valve corresponds to the working range of the fertilizer pressure. on the contrary, the working range under wide aperture of the back water valve is narrow.

Fig. 7 is the average consumption curve of the corresponding fertilizer pressure at different aperture of the back water valve degrees of liquid fertilizer. As the aperture of the back water valve becomes wider, the working range of the system is reduced. When the aperture of the back water valve is 100%, the working range has been reduced to 0 to 0.5 MPa. At this time, the system is far from being able to reach the target value within the ideal time. The efficiency is low, and the response speed is also slow. At the aperture of the back water valve of 80%, the average consumption curve of liquid fertilizer gradually moves away from the target value as the pressure of the fertilizer increase. When the aperture of the back water valve is 40%, the average consumption curve of liquid fertilizer fluctuates around the target value, under this aperture, the adjustment time of PID is short, and the work efficiency is high, and the response is fast.

The pressure of system is the most important factor affecting the flow value. Fig. 6 shows the flow curve of different depths corresponding to the pressure of the fertilizer. At the aperture of the back water valve of 40% and 80%, as the pressure of the fertilizer increases, the flow rate also increases. The slope of the flow curve, the rate at which the output flow increases, decreases as the pressure of the 48

IFAC AGRICONTROL 2019 December 4-6, 2019. Sydney, Australia

Xue Xiuyun et al. / IFAC PapersOnLine 52-30 (2019) 43–50

49

3.3.2 Fitting analysis of frequency values under different fertilizer loading pressures Fig. 9 is a frequency fitting curve corresponding to the pressure of the fertilizer at 40% aperture. Equation (3) is the model of the curve.

f = ae bx + c

(3)

In the equation, f is the frequency value , x is the pressure value.

Fig. 7. Fertilization pressure-liquid consumption fitting curve of different aperture of the back water valve The results show that the influence of liquid pump pressure on the rate of fertilization loss is more significant. The results are consistent with the conclusions obtained by Wang Jinwu et al. in the experiment of liquid fertilizer deep application of the full elliptical gear planetary system (Wang et al. 2012). 3.3 Mathematical description of deep application of precise variables

Fig.9. Fertilization pressure-frequency fitting curve of 40% aperture of the back water valve

3.3.1 Fitting analysis of flow values under different fertilizer loading pressures

Fitting the data, equation (4) are the parameters.

Fig. 8 is a fitting curve of flow corresponding to the pressure of the fertilizer at 40% aperture. Equation (1) is the preset Gaussian function model of Curve Fitting tool in MATLAB.

y = ae

  x-b 2   -     c  

 a = 0.641  b = 2.028  c = 10.85 

(4)

(1) It can be known from the fitting formula that the frequency of the inverter increases exponentially with the increase of the pressure of the fertilizer. It further proves that the excessive pressure of the system will cause the spray pump motor to overload, making the variable fertilizer system tend to be unstable.

In the equation, y is the flow value. x is the pressure value. a is the flow peak value. b is the pressure value corresponding to the flow peak value, and c is the half width information. After the integral calculation, the flow output power of the system under different pressures can be estimated. Therefore, Gaussian function model is used to approximate the data

4. CONCLUSIONS (1) A deep application system of liquid fertilizer variable was designed, and the relationship between various parameters of the system was analyzed. The mathematical model was established on the experimental data to provide reference for subsequent precise fertilization research. (2) The accuracy measurement of the variable fertilization system was performed. In the experiment, the deep application precision of the variable is up to 99.52%. The liquid fertilizer loss of single fertilization is less than 0.2litres per minute. These improve the utilization rate of liquid fertilizer, and the fertilization quality meets the agronomic requirements.

point set. Equation (2) are the parameters. Fig. 8. Fertilization pressure-flow fitting curve of 40% aperture of the back water valve

 a = 18.23  b = 2.132  c = 1.862 

(3) Analyzing the effect of fertilizer depth on system flow and frequency, the experimental results show that as the depth of fertilizer injection increases, the flow and frequency changes of the system are negligible within the system error range. The experiment only studied the changes of various parameters of the system under the deep application conditions of 10 to 50 centimeters in the citrus orchard. The

(2)

49

IFAC AGRICONTROL 2019 50 December 4-6, 2019. Sydney, Australia

Xue Xiuyun et al. / IFAC PapersOnLine 52-30 (2019) 43–50

depth of the fertilizer was only near the deep roots of the citrus fruit trees. The fertilization experiment for the deeper soil under different fertilization environments also needs to be combined with the geomagnetic spectrometer for further research.

type. Journal of Northeast Agricultural University, 48(2), 90-96. (in Chinese) Gao Zhaoxia, Xu Xuexuan, Zhao Jiaona, et al. (2014). Review on macropore flow in soil. Acta Ecologica Sinica, 34(11), 2801-2811. Ge Zhen, Xiong Qiong, Deng Minghua, et al. (2016). Optimization Design for Virtual Prototype of Variable Rate Fertilization Based on Embedded Communication System. Journal of Agricultural Mechanization Research, (3), 28-32. Gu Zukun, Yuan Jiangping, Luo Yin, et al. (2018). Research on measuring method of motor efficiency of pump system. Journal of Drainage and Irrigation Machinery Engineering,1-5. Lang Chunling, Wang Jinwu, Wang Jinfeng, et al. (2013). Variable Fertilizer Control System for Deep-fertilization Liquid Fertilizer Applicator. Transactions of the Chinese Society for Agricultural Machinery, 44(2), 43-62. Liang Chunying, Yi Shujuan. (2010). Experimental Study on Control System Performance of Variable Rate Liquid Fertilizer Applicator.Journal of Agricultural Mechanization Research, (8), 117-120. (in Chinese) Li Cuijin, Zhou Shulin, Song Lepeng. (2018). Particle Swarm Optimization of Adaptive Fuzzy PID Control for Value Cylinder Position. Journal of Agricultural Mechanization Research, (3), 244-249. (in Chinese) Mori, R., Ohdoi, K., Shimizu, H., Nakashima, H., & Miyasaka, J. (2010). Study of the Mathematical Models for Operation of Liquid Fertilizer Application. IFAC Proceedings Volumes, 43(26), 316-321. Silva, M. J., Franco, H. C. J., & Magalhães, P. S. G. (2017). Liquid fertilizer application to ratoon cane using a soil punching method. Soil and Tillage Research, 165, 279285. Wang Hongqiang, Lin Tao, Ruan Chuanbo, Rong Fengqing, Wang Zhike, & Li Haiyan. (2005). New Fertilization Equipment - Liquid Fertilizing Gun. Shanxi Fruit Tree (3), 56-56. Wang Jinwu, Liu Yahua, Wang Jinfeng, et al. (2012). Optimization Design and Experiment of Liquid-fertilizer Applying Deep-fertilization Mechanism for Planetary Elliptic Gears. Transactions of the Chinese Society for Agricultural Machinery, 43(10), 60-65+59 Wang Jinwu, Pan Zhenwei, Zhou Wenqi, et al. (2015). Design and Test of SYJ-2 Type Liquid Variable Fertilizer. Transactions of the Chinese Society for Agricultural Machinery, 46(7), 53-58. Wang Yuan, Zheng Yu, Guo Xiaoyan, et al. (2017). Development of variable deep fertilization liquid fertilizer applicator with drilling soil cave. Jiangsu Agricultural Sciences, 45(4), 177-179. (in Chinese) Zhang Ying-zi, Chen Hai-tao, Hou Shou-yin, et al. (2015). Design and Experiment of Slace Computer Control System for Applying Variable-rate Liquid Fertilizer. Journal of Northeast Agricultural University (English Edition), 22(2), 73-79. Zhao Rongyang, Wang Bin, Jiang Zhongran. (2016) Intelligent Irrigating Systems of Agriculture Based on Zigbee Technology. Journal of Agricultural Mechanization Research, (6), 244-248. (in Chinese)

(4) Analyzing the influence of pressure on system flow and frequency, the experimental results show that system pressure and the aperture of the back water valve will affect the stability of fertilization system operation. As the system pressure increases, the flow rate and frequency will increase accordingly. When the system pressure is too high, the system will not match the flow-frequency increase. As the aperture of the back water valve becomes wider, the operating range of the system is reduced, and oscillation occurs under small operating conditions. In the experiment, the changes of various parameters of the system under the aperture of the back water valve of 40% and 80% were compared. The accurate experiment of the system pressure variables still needs further research. ACKNOWLEDGEMENTS This work is supported by Guangdong Science and Technology Planning Project(2017A020208049), Characteristic Innovation Projects of Colleges and Universities in Guangdong Province (2018KTSCX020), Special Fund for Construction of Modern Agricultural Industry Technical System (CARS-27), Guangdong Provincial Special Fund For Modern Agriculture Industry Technology Innovation Teams (2019KJ108), National Undergraduate Training Program for Innovation and Entrepreneurship (201910564030). REFERENCES Chen Xiaobing, Cheng Jinhua, Chen Yinzhen, et al. (2015). Study of Soil Macropore Spatial Structure Based on Stand Spatial Stucture Analysis Method. Transactions of the Chinese Society for Agricultural Machinery, 46(11), 174-186+194. Chen Ya-yu, Huang Feng-qiu, WANG Cui-hong, et al. (2013). Progresses of Deep-applying Fertilizer Techniques. Hunan Agricultural Sciences (21), 29-33. (in Chinese). Chiang, P. N., Tong, O. Y., Chiou, C. S., Lin, Y. A., Wang, M. K., & Liu, C. C. (2016). Reclamation of zinccontaminated soil using a dissolved organic carbon solution prepared using liquid fertilizer from food-waste composting. Journal of hazardous materials, 301, 100105. Choudhury, S., Kuchhal, P., & Singh, R. (2015). ZigBee and Bluetooth network based sensory data acquisition system. Procedia Computer Science, 48, 367-372. Fan Jun, Wang Quanjiu, Wang Yuning. (2013). Simulated effects of texture and rooting depth on soil moisture sensor placement. Journal of Drainage and Irrigation Machinery Engineering, 31(01), 70-74. Feng Jinlong, Zhou Wenqi, Tang Han, et al. (2017). Design and experiment of picking hole mechanism based on deformable-elliptic gears of liquid fertilizer injection 50