基于Simulink仿真的施肥机模糊神经网络PID控制方法  

PID Control Method of Fuzzy Neural Network of Fertilizer Spreader Based on Simulink

作　　者：张荣丹 梁春英[1] 李普 王淞 邹立雯 李圳鹏 Zhang Rongdan;Liang Chunying;Li Pu;Wang Song;Zou Liwen;Li Zhenpeng(Engineering College of Heilongjiang Bayi Agricultural Reclamation University,Heilongjiang Daqing 163319)

机构地区：[1]黑龙江八一农垦大学工程学院,黑龙江大庆163319

出　　处：《南方农机》2023年第7期17-20,23,共5页

摘　　要：【目的】农用电气设备的部件十分复杂,易受到环境因素的影响,无法准确控制施肥量,故障频发。因此,需要一种成本低、运行稳定、施肥量控制精确、受工作环境影响小的施肥机控制系统。【方法】本研究采用模糊神经网络PID控制方法,将机器学习和非线性模糊控制方法引入传统PID控制器中,并基于MATLAB中Simulink仿真平台建立了模糊神经网络PID控制系统仿真模型。【结果】该模型可以根据模糊规则进行练习和学习,通过控制步进电机的转速和旋转角度来控制施肥量,神经网络还可以持续自动学习并动态改进系统的控制参数,以适应不断变化的农田作业环境。【结论】该模糊神经网络PID控制系统自我矫正能力强,可抵抗环境干扰,控制效果显著。[Objective] The components of agricultural electrical equipment are very complex and vulnerable to environmental factors, which can not accurately control the amount of fertilizer application, and frequent failures. Therefore, a fertilizer applicator control system with low cost, stable operation, precise fertilizer application control and little influence by working environment is needed. [Method] This research adopts the fuzzy neural network PID control method, introduces machine learning and nonlinear fuzzy control methods into the traditional PID controller, and establishes the fuzzy neural network PID control system simulation model based on the Simulink simulation platform in MATLAB. [Result] The model can be practiced and learned according to fuzzy rules, and the amount of fertilizer can be controlled by controlling the rotation speed and rotation angle of the stepper motor.The neural network can also continuously and automatically learn and dynamically improve the control parameters of the system,so as to adapt to the changing environment of farmland operation. [Conclusion] The fuzzy neural network PID control system has strong self-correction ability, can resist environmental interference, and has significant control effect.

关 键 词：模糊神经网络PID 步进电机 SIMULINK仿真 施肥机控制系统 

分 类 号：S224.2[农业科学—农业机械化工程] TP183[农业科学—农业工程]