基于粒子群优化RBF−PID的液肥变量施肥控制系统研究

Variable Rate Fertilization Control System for Liquid Fertilizer Based on PSO Optimized RBF−PID Control

摘要:

针对阀控液压马达液肥变量施肥控制系统稳态精度低和响应速度慢等问题，提出了一种基于粒子群优化 RBF?PID（PSO?RBF?PID） 控制的液肥变量施肥控制算法。首先建立液肥变量施肥控制系统闭环传递函数，利用粒子群算法对RBF神经网络关键参数进行寻优，并以传统PID控制和RBF?PID控制为对照，采用Matlab/Simulink软件进行仿真分析。仿真结果表明，PSO?RBF?PID控制下系统调节时间和跟踪误差均最小，验证了算法的可行性。搭建液肥变量施肥控制系统试验台架并进行室内试验，对系统流量测量精度进行验证，结果表明系统测量相对误差小于4%，满足测量要求。不同算法控制下进行系统静态和动态特性试验，试验结果表明，传统 PID控制、RBF?PID控制和 PSO?RBF?PID控制下系统流量最大相对误差分别为5.33%、3.83%、2.50%，目标流量突变时系统平均调节时间分别为5.16、3.80、2.19 s。所提出的PSO?RBF?PID控制算法各个指标均优于传统PID和RBF?PID控制算法，能够保证系统具有良好的静态和动态特性，满足液肥变量施用要求。

Abstract:

Aiming at the problems of low steady-state accuracy and slow response speed of variable fertilization control system for liquid fertilizer of valve-controlled hydraulic motor, a variable fertilization control algorithm for liquid fertilizer based on particle swarm optimization RBF?PID（PSO?RBF?PID） was proposed. Firstly, the closed-loop transfer function of the control system of liquid fertilizer variable fertilization control system was established, and the key parameters of RBF neural network were optimized by PSO algorithm, and compared with traditional PID and RBF-PID control，Matlab/Simulink software was used for simulation analysis. The simulation results showed that the adjustment time and tracking error of the system under PSO ?RBF ? PID control were the smallest，which verified the feasibility of the algorithm. A test bench of liquid fertilizer variable rate fertilization control system was built and indoor experiments were carried out to verify the flow measurement accuracy of the system. The results showed that the relative error of the system measurement was less than 4%, which met the measurement requirements. The static and dynamic characteristics of the system under the control of different algorithms were tested. The experimental results showed that the maximum relative errors of system flow under the control of traditional PID，RBF ? PID and PSO ? RBF ? PID were 5.33%，3.83% and 2.50%, respectively, and the average adjustment time of the system was 5.16 s，3.80 s and 2.19 s, respectively, when the target flow value changed suddenly. Each index of the proposed PSO?RBF?PID control algorithm was superior to that of the traditional PID and RBF ?PID control, which can ensure that the system had good static and dynamic characteristics and met the variable application requirements of liquid fertilizer.