基于迭代学习的火电单元机组协调控制研究

发布时间：2019-05-30 01:56

【摘要】：火电单元机组由锅炉、汽轮机和发电机共同组成,是一类典型的热工对象,具有大迟延、强耦合、多输入多输出等特点。由于机组动态性能受物理结构属性、运行方式及工况、外部环境等诸多因素的影响,难以建立准确的数学模型,基于模型的控制器难以达到满意的控制效果。本文主要利用单元机组大量的在线运行数据,在不需要建立其精确数学模型的情况下,利用迭代学习控制理论开展对火电厂单元机组协调控制进行研究。首先,在对国内外单元机组协调控制方案进行分析研究的基础上。分析了单元机组运行过程中的协调控制及各种工况参数数据变化的规律及重复性,采用迭代学习控制理论研究单元机组的协调控制算法。主要内容包括:1)针对单元机组建立动力学模型,并采用偏差线性化的方法进行线性化处理;2)设计了单元机组的迭代学习控制器,在理论上分析了基于压缩映射方法控制系统的收敛性;3)针对单元机组在实际过程中出现的初始条件不完全重复的现象,设计改进的迭代学习控制器,理论上给出了控制算法并进行了收敛性分析;4)通过MATLAB/SIMULINK构建仿真实验平台对所提出的算法进行了仿真验证。研究结果表明,迭代学习控制可以有效利用单元机组的重复信息,实现有限时间内的高精度协调控制。
[Abstract]:Thermal power unit is composed of boiler, steam turbine and generator. It is a typical thermal object, which has the characteristics of large delay, strong coupling, multiple input and multiple output. Because the dynamic performance of the unit is affected by many factors, such as physical structure attributes, operation mode and working conditions, external environment and so on, it is difficult to establish an accurate mathematical model, and the model-based controller is difficult to achieve satisfactory control effect. In this paper, a large number of on-line operation data of unit units are used to study the coordinated control of units and units in thermal power plants by using iterative learning control theory without the need to establish its accurate mathematical model. First of all, on the basis of the analysis and research of the coordinated control scheme of the unit at home and abroad. The law and repeatability of coordinated control and data variation of various working conditions in the operation process of unit units are analyzed, and the coordinated control algorithm of unit units is studied by using iterative learning control theory. The main contents are as follows: 1) the dynamic model is established for the unit, and the deviation linearization method is used to linearize the unit. 2) the iterative learning controller of the unit is designed, and the convergence of the control system based on compression mapping method is analyzed theoretically. 3) in view of the incomplete repetition of the initial conditions in the actual process of the unit, an improved iterative learning controller is designed, and the control algorithm is given theoretically and the convergence analysis is carried out. 4) the proposed algorithm is verified by MATLAB/SIMULINK simulation platform. The results show that the iterative learning control can effectively make use of the repetitive information of the unit and realize the high precision coordinated control in the limited time.
【学位授予单位】：河南理工大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TM621

【相似文献】