Transportation Control of Double-Pendulum Cranes With a Nonlinear Quasi-PID Scheme: Design and Experiments

Transportation Control of Double-Pendulum Cranes With a Nonlinear Quasi-PID Scheme: Design and Experiments
复制标题

采用非线性准PID方案的双摆起重机运输控制:设计与实验

DOI:
10.1109/tsmc.2018.2871627
复制
发表时间:
2019-07-01
影响因子:
8.7
通讯作者:
Chen, He
Chen, He
中科院分区:
计算机科学1区
文献类型:
--
作者:
Sun, Ning;Yang, Tong;Chen, He

文献摘要

被引文献

相似文献

在实际应用中,工业起重机通常在许多情况下遭受由所谓的双摆现象引起的影响。然而,目前在设计控制方法时,双摆现象通常被直接粗略地忽略。对于双摆起重机,目前大多数可用的方法是开环控制;现有的反馈方法大多是基于线性化的动态模型(在平衡点附近)开发的,或者在控制律中没有添加积分项,这可能会在未建模动态的存在下引起定位误差。针对这些问题,本文提出了一种新的准比例积分微分控制方法,有效地控制欠驱动双摆起重机系统。然后,我们提供了严格的理论分析的基础上,原来的非线性动力学方程的闭环系统的平衡点。据我们所知,本文给出了第一个无被控对象参数控制器,该控制器在控制器设计或闭环分析过程中不进行任何线性化操作,同时包含积分作用和执行约束,这从理论上确保了控制器在存在未建模动态(例如,不充分的摩擦补偿)、致动约束,以及大的摆动角度(即,不满足线性化条件)。最后,提供硬件实验结果来检验所建议的控制方法的有效性。
In real-world applications, industrial cranes commonly suffer from effects caused by the so-called double-pendulum phenomenon in many situations. However, at present, the double-pendulum phenomenon is usually directly roughly neglected when designing control methods. For double-pendulum cranes, most currently available approaches are open loop control; the existing feedback methods are mostly developed based on linearized dynamic models (around the equilibrium point) or designed without adding integral terms in the control laws, which may cause positioning errors in the presence of unmodeled dynamics. To address these problems, this paper proposes a new quasi-proportional integral derivative control method to effectively control underactuated double-pendulum crane systems. Then, we provide rigorous theoretical analysis for the equilibrium point of the closed-loop system based on the original nonlinear dynamic equations. To our knowledge, this paper gives the first plant-parameter-free controller that incorporates both integral action and actuating constraints without any linearizing operations during controller design or closed-loop analysis, which theoretically ensures that the controller can work well in the presence of unmodeled dynamics (e.g., insufficient friction compensation), actuating constraints, and large swing angles (i.e., not satisfying linearization conditions). Finally, hardware experimental results are provided to examine the effectiveness of the suggested control method.