Abstract:At present, the anti-dust effect is poor and the monitoring precision is low for the monitoring sensor of domestic fertilization planter. Besides it can not adjust the amount of fertilizer. A monitoring system of fertilization seeder based on Cortex-M3 processor was designed in order to settle these problems. The system could adjust the amount of fertilizer according to the working speed. It also could monitor conditions of seeding and fertilizing in time, count quantity of seeding and area statistic. When the fault occurred, it would send out alarm by buzzer and show fault type. The system adopted one piece of silicon photoelectric diode which had a large photosensitive area as receiving component and three infrared light emitting diode (LEDs) to design the seed sensor, which contained a vitreous dust cover. The structural and installation parameters of sensor were optimized to eliminate monitoring blind area. All the above helped to improve the dust-laden environmental adaptability and accuracy of monitoring system. The system adopted outer groove-wheel to design fertilizer apparatus which was driven by a worm-gear direct current(DC)motor with the rated voltage of 12V and power of 50W. Besides one end of worm shaft was connected to outer groove-wheel and the other connected to an incremental encoder for motor speed measurement. The encoder’s accuracy was 100 P/R (pulses/revolution). The main hardware circuits such as sensor signal acquisition circuit, motor drive circuit and controller area network (CAN) bus interface circuit were designed by Altium Designer software. The display interface contained three parts: main interface, setting interface and query interface. They were designed based on four rows corn planter. The seeding performance of each row was displayed in the form of a number and ring icon. The numbers showed specific seeding values. When the performance became poor, the ring icon was turned from green to red. The software development environment was Keil uVision5, and the program was written in C language. The system adopted incremental proportion integration differentiation (PID) control algorithm for the speed adjustment of fertilization motor. Lab test showed that when presetting fertilization amount was 75kg/hm2, the work breadth was 2.6m, speed was 3km/h, 4km/h, 5km/h, 6km/h and 7km/h, respectively, fertilization was varied with the speed. Fertilization deviations were 2.88%, 4.63%, 0.74%, 2.47% and 3.17%, respectively. When presetting motor rotation speed was 150r/min, seeding quantity counting deviation of the system were within 4%. The monitoring system was installed on no-tillage corn precision planter of Debont company to conduct a field trial. When presetting fertilization amount was 75kg/hm2, vehicle run 1000m with a stable speed of 3km/h, 4km/h, 5km/h,6km/h and 7km/h, respectively, the coefficient of variation of fertilization were 1.70%, 1.98%, 1.92%, 1.76% and 2.63%, respectively. Meanwhile, fertilization deviation was less than 5%. Therefore, consistency and accuracy of the system were good, which could meet the requirement of actual production. The study provided reference for the research of variable rate fertilization control technology and development of intelligent monitoring system.