基于PSO—模糊PID的变量施肥控制系统设计与试验OA
Design and experiment of variable fertilization control system based on PSO-fuzzy PID
针对槽轮式颗粒肥变量施肥系统排肥强非线性、传统控制策略适配性不足的问题,提出一种基于粒子群优化(PSO)的变量施肥控制方法.基于槽轮式排肥器的结构特性与固体颗粒肥的流动特性,引入低速极限填充率与高速衰减系数,构建考虑填充率衰减与转速饱和效应的"转速-流量"非线性排肥模型,设计"检测-计算-驱动-输出"的闭环控制框架,采用以转速误差及其变化率为输入、PID参数修正量为输出的双输入三输出模糊PID控制器,实现对PID参数的在线自适应整定.引入线性递减惯性权重的粒子群优化算法(PSO),以系统误差平方积分、调节时间ts 与超调量δ的加权和作为适应度函数,对模糊PID的初始参数进行全局寻优.结果表明,PSO—模糊PID控制系统的响应时间为0.25 s,超调量为1.3%;较传统PID控制(响应时间0.1 s,超调量43%),有效抑制了系统的剧烈振荡;较普通模糊PID控制(响应时间0.85 s,超调量10%),响应速度显著提升,综合动态性能最优.在变量作业工况下,PSO—模糊PID控制策略的施肥量平均相对误差为7.58%,最大绝对误差为4.45 g.相比于传统PID(12.83%)和模糊PID(9.91%),其平均相对误差分别减小了4.87%和1.95%.在田间作业条件下,传统 PID的平均相对误差为16.7%,模糊 PID的平均相对误差为12.68%,PSO—模糊PID的平均相对误差为10.09%.PSO—模糊PID控制系统效果综合最优,适应性更强.该控制系统精度与抗干扰性良好,提升了颗粒肥变量施肥作业的精准性与稳定性.
To overcome the strong nonlinearity of fertilizer discharge and the limited adaptability of conventional control strategies in grooved-wheel granular fertilizer variable-rate application systems,this study proposes a variable-rate fertilization control method based on particle swarm optimization(PSO).Based on the structural characteristics of the grooved wheel fertilizer dispenser and the flow characteristics of solid granular fertilizer,a low-speed limit filling rate and a high-speed attenuation coefficient are introduced to construct a nonlinear"speed-flow"fertilizer discharge model considering filling rate attenuation and speed saturation effects.A closed-loop control framework of"detection-calculation-actuation-output"is designed,employing a dual-input,three-output fuzzy PID controller with speed error and its rate of change as inputs and PID parameter corrections as outputs to achieve online adaptive tuning of the PID parameters.A particle swarm optimization(PSO)algorithm with linearly decreasing inertia weights is introduced,using the weighted sum of the integral of the squared system error(ISE),settling time(ts),and overshoot(δ)as the fitness function to globally optimize the initial parameters of the fuzzy PID controller.The results show that the PSO-fuzzy PID control system has a response time of 0.25 s and an overshoot of 1.3%.Compared with traditional PID control(response time 0.100 s,overshoot 43.0%),it effectively suppresses severe system oscillations.Compared with ordinary fuzzy PID control(response time 0.850 s,overshoot 10.0%),the response speed is significantly improved,and the overall dynamic performance is optimal.Under variable operating conditions,the average relative error of fertilizer application by the PSO-fuzzy PID control strategy is 7.96%,and the maximum absolute error is less than 4.45 g.Compared with traditional PID(12.83%)and fuzzy PID(9.91%),its average relative error is reduced by 4.87%and 1.95%,respectively.Under field operating conditions,the average relative error of traditional PID is 16.70%,the average relative error of fuzzy PID is 12.68%,and the average relative error of PSO-fuzzy PID is 10.09%.The PSO-fuzzy PID control system has the best overall performance and stronger adaptability.The proposed control system features high precision and disturbance rejection,greatly boosting the accuracy and stability of granular variable-rate fertilization operations.
杨利炜;李瑜心;王璟琪;陈浩;张雨涵;雷小龙
四川农业大学机电学院,四川 雅安,625014四川农业大学机电学院,四川 雅安,625014四川农业大学机电学院,四川 雅安,625014四川农业大学机电学院,四川 雅安,625014四川农业大学机电学院,四川 雅安,625014四川农业大学机电学院,四川 雅安,625014||四川省智慧农业工程技术研究中心,四川 雅安,625014
农业科技
固体颗粒肥变量施肥模糊PID控制粒子群优化非线性排肥模型
solid particle fertilizervariable rate applicationfuzzy PID controlparticle swarm optimizationnonlinear fertilizer discharge model
《智能化农业装备学报(中英文)》 2026 (3)
61-72,12
国家自然科学基金项目(31901413)成都市重点研发支撑计划项目(2025-YF09-00022-SN)大学生创新训练计划项目(S202510626047) National Natural Science Foundation of China(31901413)Chengdu Key Research and Development Project(2025-YF09-00022-SN)National Undergraduate Training Program for Innovation and Entrepreneurship(S202510626047)
评论