基于在线辨识与优化的自适应PID控制算法的工程实现

发布时间：2018-12-13 07:40

【摘要】：随着气候变化、环境问题日益凸显,“还城市一片蓝天”的任务也日趋严峻。为了响应节能减排降耗的目标,火电厂热工过程控制要求越来越高。面对日趋复杂的控制系统,如何实现稳定准确快速的控制成为相关学者研究的热点之一。在线辨识利用实时采集的数据不断更新被控对象模型,以便于进行更符合现场工况的控制操作。针对目前许多时变、大迟延、耦合程度高的系统,简单的PID不能完全达到控制品质要求,因此,通过基于实时数据的对象模型进行自适应PID控制,对改善系统的控制效果具有重要意义。该方法可以在线辨识与整定优化控制器参数,且具有结构简单、稳定可靠等优点。本篇论文以工程应用为目的,研究了基于在线辨识与优化的自适应PID控制算法。首先,介绍不同类型的自适应控制方法的原理和关系,并进行仿真研究。然后,介绍系统辨识常用模型及其基本的辨识方法,仿真比较发现粒子群优化算法辨识在工程中的可行性。故本文提出一种在线辨识的思想,选取符合要求的数据,辨识出不同负荷情况下,喷水量对主蒸汽温度的模型传递函数,得到模型参数与控制器参数的关系,并自动修改控制器参数,构成基于在线辨识与优化的自适应PID控制算法,仿真验证可行后。最后,利用Siemens SIMATIC S7-300 PLC实现本文的控制算法。
[Abstract]:With climate change, environmental problems become increasingly prominent, and the task of "returning a blue sky to a city" is becoming increasingly serious. In order to meet the goal of saving energy and reducing energy consumption, thermal process control in thermal power plants is becoming more and more demanding. In the face of increasingly complex control systems, how to achieve stable, accurate and fast control has become one of the hot research topics. On-line identification uses real-time data to update the model of the controlled object, so as to facilitate the control operation which is more in line with the field operation conditions. For many systems with time-varying, large delay and high coupling degree, simple PID can not fully meet the control quality requirements. Therefore, adaptive PID control is implemented through an object model based on real-time data. It is of great significance to improve the control effect of the system. This method can identify and adjust the controller parameters on line, and has the advantages of simple structure, stability and reliability. For the purpose of engineering application, an adaptive PID control algorithm based on online identification and optimization is studied in this paper. Firstly, the principle and relation of different adaptive control methods are introduced and simulated. Then, the common models of system identification and their basic identification methods are introduced. The simulation results show that particle swarm optimization algorithm is feasible in engineering. In this paper, an idea of on-line identification is proposed. The model transfer function of water injection amount to the main steam temperature is identified under different loads, and the relationship between the model parameters and the controller parameters is obtained by selecting the data in accordance with the requirements. The adaptive PID control algorithm based on on-line identification and optimization is constructed by modifying the controller parameters automatically. The simulation results show that the algorithm is feasible. Finally, Siemens SIMATIC S7-300 PLC is used to realize the control algorithm.
【学位授予单位】：华北电力大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP273

【参考文献】