【作者】 张勇；

【导师】 穆朝絮； 王伟；

【作者基本信息】 天津大学 ， 控制工程， 2019， 硕士

【摘要】 随着工业生产水平的提高,相关工业过程的控制精度要求也逐渐增高。然而,实际生产中存在的各种不确定性与扰动将直接或间接影响控制精度,进而影响产品的最终品质。本文以自适应动态规划相关控制技术理论为基础,针对非线性系统在各种不确定性和干扰影响下的鲁棒控制问题进行研究。文章主要从理论研究和相关应用场景两个方面进行展开,并针对具体场景进行分析与设计。本文首先考虑了不确定非线性系统的鲁棒跟踪控制问题。考虑非线性系统含有的匹配/非匹配不确定性,并给出代价函数设计思路。进一步,针对含有非匹配不确定性的非线性系统鲁棒跟踪控制问题进行具体分析。建立辅助系统处理非匹配不确定性,通过稳态与暂态分析,得到暂态跟踪误差系统,并设计鲁棒跟踪控制策略。结合自适应动态规划方法,最终得到跟踪控制问题的近似最优解。同时,提供了基于李雅普诺夫稳定性分析的理论证明,并辅以数值仿真对其有效性进行了验证。以下分别以四旋翼飞行器和微电网系统为背景,考虑不确定问题并设计自适应鲁棒控制策略。考虑四旋翼飞行器系统中的时变与耦合不确定问题,设计一种基于神经迭代算法的鲁棒跟踪控制策略。基于位置和姿态跟踪误差子系统,提出一种改进的神经网络权值更新规则,放松对初始稳定控制策略的要求,进而通过迭代算法得到近似最优控制律。最后,考虑时变、耦合不确定问题,给出基于线性和非线性四旋翼飞行器系统的两个仿真算例,用以验证本文控制策略的有效性。针对微电网系统,随着可再生能源发电比例的增加,随机发电和动态不确定性为微电网频率控制带来了极大的挑战。本文提出一种完全基于自适应动态规划的智能频率控制策略,用于调节微型燃气轮机和电储能系统的功率输出。此外,考虑一天中光伏发电和负荷需求的变化,通过设计储能规划策略,可以实现光伏发电的充分利用和电储能的循环使用。仿真算例对基于电储能系统的智能频率控制策略进行了验证,并通过与PID、线性二次调节器和模糊控制方法的对比,体现了本文控制策略的优越性。更多还原

【Abstract】 With the improvement of industrial production level,the requirement for control precision of related industrial processes is gradually increasing.However,various uncertainties and disturbances exist in the actual production which will directly or indirectly affect the control precision,and then affect the final quality of the product.Based on the theory of ADP,this paper studies the robust control of nonlinear systems under various uncertainties and disturbances.This paper mainly develops from two aspects of theoretical research and related application scenarios,and carries out specific analysis and design for specific scenarios.Firstly,considering the robust tracking control problem of uncertain nonlinear systems,the matched/unmatched uncertainties are analyzed and the design idea of corresponding cost function is given.Further,the robust tracking control problem of nonlinear systems with considering unmatched uncertainties is analyzed.An auxiliary system is estabilished for the unmatched uncertainties.Then,a transient tracking error system is obtained through steady and transient analysis,and the robust tracking control strategy can be obtained.Based on the ADP framework,the approximate optimal solution of the robust tracking control problem is obtained.Meanwhile,the robust tracking control strategy is theoretically guaranteed by using Lyapunov stability analysis,and the effectiveness is verified by simulation experiments.In the following,quadrotor aircraft and microgrid system are used as backgrounds to consider uncertainties and design adaptive robust control strategies.For time-varying and coupling uncertainties in a quadrotor aircraft system,a neural-iterative-based robust tracking control strategy is designed.Based on the position and attitude tracking error subsystem,an improved weight updating rule of neural network is proposed to relax the requirement of initial stability control,and an approximate optimal control law is obtained through an iterative algorithm.Finally,considering the time-varying and coupling uncertainties in the actual flight environment,three simulation cases based on linear and nonlinear quadrotor dynamics are given to demonstrate the effectiveness of the control strategy.For the microgrid systems,with the increasing proportion of renewable energy generation,stochastic power generation and dynamic uncertainty bring great challenges to the frequency control of microgrid.In this paper,an intelligent frequency control strategy based on ADP is proposed to adjust the power output of micro-turbine and energy storage system(ESS).In addition,considering the changes of PV power and load demand in a day,the full utilization of PV power and the recycling of ESS can be realized through the proposed regulation strategy.Simulation cases validate the energy-storage-based intelligent frequency control strategy for the microgrid.Compared with PID,fuzzy logic control and linear quadratic regulator(LQR),the superiority of the proposed control strategy is demonstrated.更多还原