Fertilization Control System Research in Orchard Based on the PSO-BP-PID Control Algorithm

In order to improve the precision of the variable-rate fertilization system in orchards, this paper conducted a simulation by MATLAB and experimental research based on a variable-rate fertilization experiment platform. The variable-rate fertilization experimental platform was mainly composed of a power supply, DC motors, a PPC-15A1 on-board computer that contains a PCI8932 PC-DAQ, speed sensors, fertilizer dischargers, and a NAV60 module that can receive Beidou Navigation Satellite System positioning data. According to the fertilizer application mechanism of an external grooved wheel fertilizer applicator, the control system model of the variable-rate fertilization driven by the DC motor for orchards was established. A BP neural network adaptive PID controller based on particle swarm optimization (PSO) was proposed to improve the control precision of the system. The step response simulation results by MATLAB show that the overshoot of the BP-PID controller optimized by the PSO algorithm (PSO-BP-PID) is 12.7%, and the adjustment time is 0.557 s. The variable-rate fertilization experiments were conducted, in which the control system was tested by using the PSO-BP-PID controller. The variable fertilizer seeder control system of the Chinese national standard was adopted to evaluate the performance indexes of the system, such as the range of fertilizer amount adjustment, the response time of fertilizer amount adjustment, and the control precision of fertilizer amount. In the variable rate fertilization experiments, the average fertilization errors, respectively, are 1.16% and 1.07%, under the conditions of changing the target fertilization amount and the vehicle speed. The test results are consistent with the simulation results, and the variable-rate fertilization performance parameters are improved.

Automated summary

This paper improves the precision of variable-rate fertilization in orchards by conducting simulations and experimental research. A BP neural network adaptive PID controller based on particle swarm optimization is proposed to enhance the control precision of the system. Experimental results show improved variable-rate fertilization performance parameters.