Research On The Real-time Predictive Control System Of Variable Fertilization In Corn Field Cultivation Based On NDVI

Variable fertilization technology based on crop nutrient information is one of the important directions of future agriculture.Its technology makes up for insufficient and uneven fertilization in the early stage based on crop nutrition information.In the past,variable fertilization information processing was mostly in the form of prescription maps,which lacked real-time processing and application of information..Due to the lack of real-time processing methods,the real-time performance of sensor variable fertilization cannot be guaranteed,so the agronomic requirements and technical guidance of variable fertilization in corn field cultivation are combined.Aiming at the accuracy and real-time problem of fertilization position of variable fertilization,a rapid variable fertilization control system was developed,and the control effect of the system was tested and verified.(1)According to the structural characteristics of the corn fertilizer applicator,the design target of the variable control system was analyzed.The actual working conditions and the agronomic requirements of cultivating and topdressing were combined.The system hardware equipment was mounted on the basic structure of variable fertilization machinery.Active remote sensing spectroscopy testing technology was systematically used to determine the amount of topdressing for nitrogen nutrition diagnosis.CAN-bus communication was used to transfer data information and complete real-time sharing of information between devices.(2)Based on the spatial correlation of maize canopy NDVI values,the correlation trend changes in the spatial range were verified,and the predictive control theory was proposed.The adaptive cubic index prediction algorithm was selected to predict the NDVI numerical value to achieve data filtering and trend prediction.The predicted NDVI data was combined with the fertilization optimization algorithm,and the next stage of fertilization will be adjusted in advance.The error of the prediction algorithm was verified and analyzed,and the error range was within9.8% to ensure the stability and reliability of the variable control effect.(3)The influencing factors of the delay and lag of the variable fertilization process were analyzed,and the integral separation control algorithm was introduced to reduce the overshoot phenomenon.In order to ensure the accuracy of the fertilization location,the lag correction model was used to modify its control system,and the threshold control method was used to calculate the collection point and the location of fertilization in real time.The variable fertilization software was designed according to the lag correction model and prediction algorithm.The program was developed according to actual functional requirements.Combined with the simulation analysis of the lag correction effect,the response jump delay time was 0.8 s,and the result shows that the minimum fertilization position was 0.2 m,which achieves the purpose of system control and real-time monitoring.(4)The indoor bench test was used to analyze the accuracy of fertilization,and the actual field variable fertilization was used to verify the practical performance of the system.The system response characteristic test shows that setting three different speeds under load conditions,the system’s start-up response time was 1.1 s,the overshoot was 8.1%,but when the speed jump change time was 0.8 s,the absolute value of the speed dynamic error was 0.85 r /min.The average relative error between the theoretical rotational speed of the hydraulic drive and the actual monitored rotational speed of the actual field operation is 1.35%.The variable fertilization control system meets the real-time nature and realizes the requirements of precise variable fertilization in the region.