English

Journal of Chinese Agricultural Mechanization

Journal of Chinese Agricultural Mechanization ›› 2021, Vol. 42 ›› Issue (9): 83-89.DOI: 10.13733/j.jcam.issn.2095-5553.2021.09.12

Previous Articles     Next Articles

Research on water and fertilizer control system based on PSOBP optimization PID model

Song Zhuoyan, Xu Xiaohui, Song Tao, Cui Yinggang, Si Yulong.    

  • Online:2021-09-15 Published:2021-09-15

基于PSO-BP优化PID模型的水肥控制系统研究

宋卓研;徐晓辉;宋涛;崔迎港;司玉龙;   

  1. 河北工业大学电子信息工程学院;
  • 基金资助:
    河北省重点研发计划项目(19227212D,20327201D)
    石家庄市重点研发计划项目(191490144A)

Abstract:  Aiming at the problems of traditional irrigation and fertilization methods that cannot meet crop growth requirements coupled with the serious waste of water and fertilizer resources, a water and fertilizer control system based on PSO and BP neural network to optimize the PID model was designed. By combining the water and fertilizer concentration information of the crop planting environment, the system used the PSO and BP neural network algorithm to optimize PID control parameters to solve the problems of nonlinear, timevarying, and hysteresis in the water and fertilizer irrigation process. The MATLAB/Simulink simulation test results showed that the PID control model optimized by PSO and BP neural network was 9.33% faster than the traditional PID control model, the adjustment time was shortened by 72.24%, and the overshoot was only 11.78% of the PID control. The optimization effect was good. The system test results showed that the system control was stable during the irrigation process, and the effect of precise control of water and fertilizer concentration was achieved to a certain extent, which had practical application value.

Key words: water and fertilizer control, PID control, PSO optimization, BP neural network, MATLAB simulation

摘要: 针对传统灌溉施肥方式无法切实满足作物生长需求和水肥资源浪费严重的问题,设计一种基于PSO和BP神经网络优化PID模型的水肥控制系统。系统通过结合作物种植环境水肥浓度信息,利用PSO和BP神经网络算法优化PID控制参数,以解决水肥施灌过程中系统的非线性、时变性和滞后性等问题。综合MATLAB/simulink仿真试验结果可知,利用PSO和BP神经网络优化的PID控制模型较传统PID控制模型系统响应速度提高9.33%,调节时间缩短72.24%,超调量仅为PID控制的11.78%,优化效果较好。系统试验结果表明,施灌过程中系统控制稳定,在一定程度上达到水肥浓度精准控制的效果,具有实际应用价值。

关键词: 水肥控制, PID控制, PSO优化, BP神经网络, MATLAB仿真

CLC Number: