基于模糊PID算法的果树精准喷施变量控制系统研究

Variable Control Technology for Precision Spraying of Fruit Trees Based on Improved Fuzzy PID Algorithm

摘要:

针对果园喷雾机在小流量喷施区间内控制精度不足、稳定性差等问题，本文设计了基于模糊PID算法的果树精准喷施变量控制系统。通过PWM信号控制直流无刷水泵转速精确控制流量，通过试验得到不同占空比与直流无刷水泵流量关系，并根据数据变化趋势进行分段拟合，得到精确的流量拟合公式。将模糊控制与传统PID控制算法结合，采用高斯函数和非对称三角函数设计隶属度函数，实现了模糊PID算法。为验证算法有效性，对传统PID、模糊控制与本文算法进行了建模仿真，试验结果表明，本文算法在上升时间、超调量、稳态误差方面均优于其他算法。同时，设置了非运动状态下流量控制精度、运动状态下动态流量跟随试验及液滴沉积率测试。试验结果表明，本文算法喷施量误差控制在2%以内;当喷施量在0.5~3 L/min范围内动态变化时，平均超调量为2.43%，平均调整时间为0.52 s，液滴平均沉积率为76.89%。研究结果为智慧果园发展和应用提供了技术支持。

Abstract:

Aiming at the problems of insufficient control accuracy and poor stability of the orchard sprayer in the small flow spraying interval, a variable control system for precision spraying of fruit trees was designed based on the improved fuzzy proportional-integral-derivative (PID) algorithm. The system controlled the rotational speed of the DC brushless water pump through the pulse width modulation (PWM) signal to control the flow rate accurately. The flow rate relationship of the DC brushless water pump under different duty cycles was obtained through the test, and the segment fitting was performed according to the trend of the data to obtain the accurate flow rate fitting formula. Then the fuzzy control was combined with the traditional PID control algorithm, and the Gaussian function and asymmetric trigonometric function were used to design the affiliation function, which achieved the improvement of the fuzzy PID algorithm. In order to verify the effectiveness of the algorithm, the modelling simulation comparison between the traditional PID, fuzzy control and the present algorithm was carried out, and the experimental results showed that the present algorithm was better than the other algorithms in terms of rise time, response overshooting amount, and steady state error. At the same time, the comparison experiments of flow control accuracy in non-motion state, dynamic flow following in motion state and the test of droplet deposition rate were set up. The experimental results showed that the spray volume error of this algorithm was controlled within 2%;when the spray volume changed dynamically in the range of 0.5~3 L/min, the average overshooting amount was 2.43%, the average adjustment time was 0.52 s, and the average droplet deposition rate was 76.89%. The research result can provide technical support for the development and application of smart orchards.