| China's facility agriculture has developed rapidly.The total area of facility agriculture ranks first in the world.Facility agriculture has played a great role in enriching people's dining table and promoting agricultural economic development.However,in facility agriculture,due to the large output of crops and the high consumption of fertilizer,in order to avoid the nutrient deficiency of crops,producers and operators commonly over-apply nitrogen,phosphorus,and potassium fertilizers.It not only leads to the waste of fertilizer,but also causes environmental non-point source pollution,resulting in crop quality decline and yield reduction.Accurate management of water and fertilizer requires the nutrient status detection of crop.Traditional nutrient detection methods need destructive sampling of crops,which affects crop growth.Sampling,determination,and data analysis of them cost a lot of manpower and material resources,and the timeliness is poor.Optical or spectral detection methods can only detect the leaf surface but cannot detect the nutrient ion concentration in crops directly.Therefore,ion selective microelectrode technology was developed to detect the nutrient concentration inside of crops directly.However,the measured nutrient ion concentration in crops by this method is only the nutrient state at that time,which cannot fully reflect the growth state of crops.Therefore,it is also necessary to consider the growth information of crops so as to reflect the real water and fertilizer demand of crops comprehensively.Aiming to solve the above problems,this dissertation chooses tomato,a typical fruit vegetable cultivated in greenhouse,as the object,and puts forward an accurate fertilization method to determine the water and fertilizer requirements and formula of tomatoes by creating a multi-barrelled ion selective microelectrode(MISME)to detect the nutrient ion concentration in plants,and combining with the data of stem diameter,growth information and fluorescence parameters of crop,and environmental factors.After overcoming these key technologies,a precise control system is developed.The main contents and innovations in this dissertation are as follow:1.A MISME was developed for the detection of nutrient ion concentration in tomato leaves.The suitable depth of MISME's penetration into tomato leaves ranged from 25.60?m to 35.59?m.The deformation of microelectrodes with different tip length and size during microelectrode puncture was analyzed,and the optimized tip length and size range of microelectrodes were determined.The ion sensitizers of nitrate ion,ammonium ion and potassium ion were selected,and the technological parameters of drawing microelectrode were obtained.The analytical model of the output signal of MISME and the concentration of nitrate,ammonium and potassium ions was established,and the relative error of detection was 1.83%,2.12% and 2.84%,respectively.2.A method for detecting nutrient ion concentration in tomato plant was proposed based on MISME.The effects of nitrate nitrogen,ammonium nitrogen and potassium in nutrient solution and their interactions on nitrogen and potassium contents in tomato plants were studied.From the third branch leaf,started from the top growth point of tomato plant to the stem,to the seventh branch leaf,each branch leaf was divided into three parts,marked as upper-leaf,middle-leaf and lower-leaf,by proportional to the length.A special feather leaf was selected randomly in each part.The special feather leaf was divided into three areas by radial equal area,marked as outer-area,inner-area,and center-area.The SPAD values of all areas were measured.The correlation coefficient between the average SPAD value of each branch and the nitrogen content of tomato plants were analyzed.The nutrient detection positions of tomato plant were determined to be the center-area of middle-leaf and the middle-area of upper-leaf of the sixth branch leaf.A model for detecting nitrogen and potassium content of tomato was established.The test results showed that the average relative errors of nitrogen and potassium content were 3.32% and 3.44%,respectively.3.A modeling method of tomato fertilizer requirement model based on nutrient ions and growth information was proposed.The influence of climate and environmental parameters on fertilizer needs of tomato was analyzed,and the prediction model of environmental parameters based on time series theory was established.The relative error of temperature prediction was 1.1%.A prediction model of stem diameter growth rate was established based on diurnal physiological development effects during vegetative growth period of tomato.The relative error of prediction was 3.5%.The effects of nutrient solution formulation on the fluorescence parameters of tomato leaves were investigated.Linear electron flow,chlorophyll fluorescence decline rate and proton outflow rate were selected as the rapid predictive values of nitrate,ammonium,and potassium ion concentrations in nutrient solution,respectively,and a regression prediction model was established.The effects of nutrient solution formulation on tomato quality and yield were explored.Four kinds of nutrient solution formulation modification methods were proposed according to nutrient ion concentration information,fluorescence parameters,expected quality and yield.The effects of irrigation amount and fertilization amount on tomato yield,water use efficiency and fertilizer were analyzed through water and fertilizer coupling experiments.The basic irrigation amount and fertilizer amount for tomato growth were determined.The difference between the actual growth rate of plant height and stem diameter and the standard growth rate of plant height and stem diameter was used to characterize the water and fertilizer deficiency of tomato.The standard growth rate model of tomato was obtained by using tomato accumulated temperature.The prediction models of tomato fertilizer and water requirement were established.4.A precise deployment method of multi-channel nutrient solution was proposed.After analyzing the advantages and disadvantages of three common methods of adding mother liquor such as differential pressure,suction and injection,the suction method with a simple structure and low cost was selected to add mother liquor in the system.The relationships between the amount of single channel and multiple channels mother liquor and values of EC and p H were studied.The regression equations of EC value and p H value of nutrient solution were established.The mean relative error of EC value of regression model was 1.64%,and the maximum was 2.75%.The average relative error of p H value was 1.76% and the maximum was 2.85%.A method for regulating the ratio of mother liquor was proposed.The EC and p H values of the nutrient liquor were converted into one-variable functions of PWM duty cycle when the nutrient liquor formula was given.The correction model of mother liquor ratio error was established,and the maximum ratio error was less than 3%.The fuzzy PID control algorithm fusion of single P control and the pre-estimation method were used to design the EC value and p H value controller of nutrient solution,and the control rules of EC value and p H value of nutrient solution were formulated.The control errors of EC and p H value under different irrigation flows were 0.08 m S/cm and 0.06,respectively.5.A precise control system of greenhouse nutrient solution was developed based on nutrient ion concentration and growth information.The main performance and technical parameters of the system were determined,and the operation process of the whole system was formulated.The system structure and the hardware and software of the control system were designed,and the prototype of WGF-6-12 automatic fertigation machine of greenhouse was manufactured.The control errors of EC value,p H value and irrigation amount were 0.05 m S/cm,0.01 and 1.0%,respectively,and the error of nutrient element ratio were between 0.6% and 2.8%.The whole machine performance test datum showed that,in comparison with the experimental group EP2 and the control group CP treatment,the tomato yield of the experimental group EP1 treatment increased by 12.16% and 8.13%,the water use efficiency increased by 35.07% and 20.84%,the fertilizer use efficiency increased by 30.25% and decreased by 0.80%,the economic benefit of single tomato plant increased by 16.52% and 18.06%,and the total sugar content increased by 8.40% and 16.35%,The total acid content decreased by 9.76% and13.95%,and the vitamin C content increased by 4.48% and 12.93%,respectively.Through the above research,the relationship between MISME signals and content of nitrogen and potassium in tomato plants is obtained.It can quickly and immediately detect the nitrogen and potassium nutrient status of tomato plants,and accurately regulate water and fertilizer according to the information such as nutrient ion concentration,growth information and fluorescence parameters.It provides a nutrient detection and fertigation decision-making method for water and fertilizer regulation,and it is of great significance to improve the economic benefits of greenhouse production and promote the scientific and technological progress of facility agriculture. |
PDF Full Text Request