| In the recent decades,due to its convenience,non-destructive characteristics,the use of spectrometry to determine the nutrient status of nitrogen in crops has become a hot topic in the research of nitrogen nutrition diagnosis.At the same time,the deep integration of smart sensor technology,intelligent information processing technology and network communication technology have greatly promoted the rapid rise of the mobile Internet,Internet of things,and cloud computing,and the rapid spread and widespread use of various mobile smart terminals.Modern agriculture badly needs the Internet of Things and cloud computing technologies to improve agricultural production efficiency.Due to the large-scale application of drip irrigation,water and fertilizer integration model are put into practice for cotton plantation in Xinjiang.At present,automatic control has been put in place to irrigation in cotton planting in Xinjiang,but fertilization still mainly relies on labor experience.With the development of the Internet of Things,the technology to incorporate fertilization information collection,decision-making,and control emerges as the times require.The precision fertilization control technology can give fertilization amount and fertilization ratio by taking into consideration soil conditions,weather and climate,plant growth and fertilizer requirements,combined with the advanced experience of crop planting experts.In recent years,considerable research has been done on the application of the Internet of Things technology in smart fertilization,but most of the studies concentrate on facility agriculture,and the application of the Internet of Things technology in fields advances slowly.The cotton planting area in Xinjiang is large,and excessive fertilization every year results in serious pollution and waste.The study adopted the Internet of Things-based smart dripping irrigation fertilization technology.The working principles of the technology function as follows.First,information of cotton nitrogen is acquired from sensors,and then transmitted to the network,and finally processed by supporting sublevel,and sent to the decision-making level.By doing so,the goal can be achieved to control the allotment and fertilization in a closed loop and promote the zero growth of national fertilization.The main findings are as follows:(1)Spectral reflectance of cotton canopy and chlorophyll value of cotton leaf were measured by ASD spectrometer,and the correlation between the vegetation index of the first derivative of the spectrum and the original spectral reflectance and the chlorophyll content of the leaf was estimated.The results showed that the maximum first-order differential value(Dr)within the red edge and Base value modified normalized difference index(Bm SR705)had a good correlation with the chlorophyll content of cotton.,Estimation model decision coefficient R2 = 0.8157**,root mean square error RMSE=0.278.Therefore,the use of specific vegetation index and "red edge parameters" can predict the chlorophyll content well,so as to provide the basis for hypersecretion data to predict the chlorophyll content of cotton leaves.(2)By setting up different combinations of water and nitrogen,we study the relationship between leaf nitrogen content and hypersecretion reflectivity under different water and nitrogen conditions,and confirm the sensitive band for the diagnosis of cotton N content,and construct a chlorophyll spectrum diagnostic model for cotton leaves.As a result,it was found that in the visible light band,the spectral reflectance decreased with the increase of the amount of used water and nitrogen.In the near-infrared band,the reflectivity increases with the increase of water and nitrogen application.The vegetation index based on m ND705,CCI,Dr and SDr/SDb can well monitor the nitrogen content of cotton leaves.In addition,when the amount of fertilizer is increased to a certain amount,the yield will no longer increase,but will be inhibited by nitrogen.Through field verification,the recommended fertilization based on the m ND705 fertilization model can reduce the total amount of fertilization and improve the efficiency of fertilization and agronomic efficiency.(3)In the aspect of target allocation by wireless sensor networks,the advantages and disadvantages of improved niche genetic algorithm,quantum evolutionary algorithm and particle swarm optimization are compared.Since the proposed distribution scheme based on improved niche genetic algorithm considers not only the advantages of sensor location but also residual energy and other factors,the parameters can be adjusted for adaptation during the operation.When the number of sensors approximate or is less than 200,the performance of quantum evolutionary algorithm and particle swarm optimization is similar to the improved niche chaotic genetic algorithm.However,when the number of sensors is more than 400,the improved niche chaotic genetic algorithm enjoys obvious advantages.(4)Based on the analysis of the advantages and disadvantages of various intelligent PID control algorithms,a self-adaptive fuzzy control algorithm was studied,which was suitable for the developed intelligent fertilizer control device,and a theoretical model of allotment and fertilization motor was established.Compared with the transient response of the conventional PID controller,the rise time is 1.2s,the overshoot is ?=42%,and when the fuzzy adaptive PID controller was used,the rise time is 0.7s and the overshoot is ?=27%.The fuzzy adaptive PID control method has better dynamic performance so that the balance can be reached more quickly and the adverse effect of excessive overshoot on the circuit can be effectively reduced.(5)Research and design of precision fertilization control devices,include various control modes,and remote connection is realized through GPRS.The smart fertilization intelligent fertilization control device in real time sends the collected nitrogen data of cotton to the smart fertilization cloud platform through GPRS module.The single batch will be divided by the time needed for allotment of fertilizer and the amount of fertilizer per time unit.The fertilized water concentration in the tank and the motor speed were analyzed,and the obtained error analysis shows a better quality than previous allotment and fertilization devices.The uniformity of the fertilizer is good and the device is very smart the uniformity of the fertilizer is good and the device is very smart and nitrogen fertilizer utilization is high.(6)A cloud platform for fertilization has been developed based on Bootstrap and Ajax technology so that it can be adapted to mobile phones and computers with different operating systems and resolutions.It is also convenient for platform expansion,spreading and application.As a result,the staff for departments of agricultural management and farmers can gain access to real-time information on crop fertilization by means of smart phones or computers and then offer advice on rational control over fertilization for more convenience and precision. |
PDF Full Text Request