【作者】 雷正玲；

【导师】 郭晨；

【作者基本信息】 大连海事大学 ， 交通信息工程及控制， 2014， 博士

【摘要】 本文针对船舶动力定位控制系统展开了仿真建模与自抗扰控制的研究,主要的研究工作包括以下四个方面：1.基于实船数据建立了动力定位控制系统的仿真模型,包括系统动力学模型、实时推力分配逻辑、推进器模型以及传感器噪声模型,使得控制器的输出量能实时转化为实际工程中的控制量——螺距角和舵角,并对所建立的系统模型进行了仿真验证。2.作者从自抗扰控制应用研究的核心问题——扰动估计切入,首先详细分析了船舶动力定位系统中所面临的各种扰动,并将其分为低频扰动和高频扰动两类。针对低频扰动,分别对速度矢量可测量和速度矢量不可测量的情况给出了相应的抗扰控制方法；针对高频扰动,探索应用了两种自抗扰的滤波方法,然后对所提出的控制和滤波方法进行了仿真验证。最后对动力定位自抗扰控制方案效果的影响因素进行了讨论分析,为后续研究奠定了基础。3.提高扩张状态观测器的扰动估计精度是提高自抗扰控制性能的重要途径,针对工程实际中可能存在的情况：1)可实现观测器带宽较小；2)采样间隔大,分别给出了相应的扰动估计模型的改进方法：1)基于双ESO的扰动估计模型；2)基于高阶扰动信息的扰动估计模型,并以锚泊辅助动力定位问题为例,仿真验证了上述改进方法的有效性。4.针对船舶动力定位控制系统中由执行机构引起的时滞问题,本文提出一种基于预估器的采用高阶扰动信息扰动估计模型的自抗扰控制器,并以动力定位系统中第三个回路为例,分别对不考虑与其它回路耦合影响和考虑与其它回路耦合影响的情形进行了仿真计算,验证了上述自抗扰控制器对于系统含有时滞的船舶动力定位控制的有效性。更多还原

【Abstract】 This paper focused on simulation modeling and Active Disturbance Rejection Control (ADRC) of ship Dynamic Position System (DPS), the related four research tasks are listed as follows:1. Simulation model of DPS was established according to the data of a real ship, including system dynamic model, real-time thrust allocation logic, thrust model and noise model of sensors, which helped transform controller outputs to engineering controlled variables, pitch angle and rudder angle, online. Simulation tests of the established system were also conducted.2. As the key point of ADRC is disturbance estimation, all kinds of disturbance being exerted on the ship were analyzed firstly and classified into low-frequency disturbance and high-frequency disturbance. To counteract the low-frequency part, two different disturbance rejection control methods were proposed according to different conditions-(a) when the speed is measurable and (b) when the speed is immeasurable; to cancel the high-frequency part, two kinds of filters of Active Disturbance Rejection control were introduced. Simulation tests of the above solutions were also carried out. Thereafter, discussions on the influential factors of the ADRC’s performance were carried out, which provided a solid foundation for the following study.3. An important way to improve ADRC’s performance is to enhance the disturbance estimation accuracy of Extended State Observer (ESO), so two methods,1) Double-ESO based disturbance estimation model and2) Disturbance estimation model based on higher order information of disturbance, were proposed separately to solve two engineering problems in DPS,1) The achievable bandwidth of ESO is relatively low and2) The sampling interval is relatively large. Simulation tests on the problem of thruster-assisted position mooring control verified the proposed methods.4. Concerning the delay produced by the thruster system of the DPS, an ADR controller based on a model-free predictor was proposed, the third loop of the DPS was taken as an example for simulation test. The impact of the coupling influences with other loops was covered in the simulation tests which helped verify the proposed method for DPS with delay.更多还原