Design And Experiment Of Cotton Precision Seeding Monitoring System For Yangtze River Basin

In recent years,China has gradually resumed production of inland cotton.However,the process of planting cotton often produces serious omissions,missing seeds and broken strips to miss the best sowing time.Difficulties in monitoring sowing,omission and reseeding during cotton precision seeding in the Yangtze River Basin need to be solved urgently.On the one hand,combined with the characteristics of cotton planting in the Yangtze River basin,the combination of LED line array light source and thin surface narrow slit structure produces a thin surface light layer to design a cotton precision seeding detection device suitable for post-wheat/oil direct seeding cotton planting pattern.On the other hand,a cotton precision seeding monitoring terminal is provided to monitor cotton sowing information in real time.(1)The cotton precision seeding detection device is designed.Based on the physical characteristics of cotton seeds,a detection method is designed by combining LED line array light source and thin surface narrow slit structure to produce a thin parallel light layer,and convex lens to converge light to silicon photocell.To ensure that the critical dimensions such as the detection tube diameter,the size of the thin narrow slit used to create thin surface light and the focal length of plano-convex lens are all matched,the main structural features of the detection device and the characteristics of the optical path at the extreme position are analyzed.And the selection of key sensing components is completed accordingly.To convert the signal into a square wave that can be recognized by a microcontroller,a signal collection circuit is provided.Meanwhile,the signal response characteristics of different signal conditioning links of cotton seeds are analyzed.(2)A cotton precision sowing monitoring terminal is designed,which can be used with cotton precision seeding detection device and millimeter wave radar to form a cotton precision seeding monitoring system.A monitoring terminal that provides real-time monitoring and display of cotton sowing quality information is designed.The system can automatically upload the sowing data to the cloud storage backup and meet the demand of downloading and querying the historical sowing information at any time.The hardware circuit design of the monitoring terminal is also carried out and the hardware components are introduced.Based on the analysis of the main functional requirements of the monitoring terminal,the functions,operation mechanism and execution flow of the software subroutines of the monitoring terminal are designed respectively.(3)Bench test and field test are conducted for the whole set of cotton precision seeding monitoring system.Firstly,the bench test is carried out,the results indicate a seeding detection accuracy of not below 93.56% when the seeding frequency is 5.87Hz～24.05 Hz.The maximum error of the sowing leakage index detection value is 2.94 percentage points when no serious sowing leakage occur,and the maximum error of the reseeding index detection value is 5.96 percentage points when no serious reseeding occur.At the same time,the monitoring system is well adapted to different seed dispensers,and the seed coating agent has no significant effect on the detection accuracy.Based on this,the monitoring error influencing factors are analyzed to further optimize the performance.In order to further improve the accuracy of the monitoring system,a sowing compensation algorithm and a missed sowing detection algorithm using the Pauta criterion have been developed.Seed rate detection accuracy is tested at 97.95% at a seeding frequency of no higher than 25.05 Hz,and the accuracy of the missed seeding detection algorithm is optimised to increase by approximately 8 percentage points,with a maximum error of no more than 19.06 percentage points.Finally,the results of the field tests conducted show that light,vibration and dust have no effect on the operation of the cotton precision seeding monitoring system.The accuracy of sowing detection reaches 96.61% when the seeding frequency is6.04Hz～17.72 Hz.Moreover,the data transmitted by the monitoring system and Alibaba Cloud platform are accurate,complete and stable,and the average waiting time for historical query operation is about 1.3s.