| 一种神经网络预测控制在超临界主汽温度中的应用研究 |
| |
| 引用本文： | 李云娟,方彦军.一种神经网络预测控制在超临界主汽温度中的应用研究[J].昆明师范高等专科学校学报,2010,32(3):97-99,103. |
| |
| 作者姓名： | 李云娟 方彦军 |
| |
| 作者单位： | [1]昆明学院自动控制与机械工程系,云南昆明650118 [2]武汉大学自动化系,湖北武汉430072 |
| |
| 摘 要： | 传统PID控制难以在非线性、迟延、时变和具有扰动特质的超临界主汽温度控制系统中达到满意的控制效果.因此,提出了一种采用多步预测、滚动优化和反馈校正的神经网络预测控制系统.以某超临界电厂主汽温度为研究对象,MATLAB仿真结果表明：不同的工况建立的主汽温度神经网络动态模型,能够很好地预测对象的动态特性,取得了优于传统PID的控制效果.
|
| 关 键 词： | 神经网络 预测控制 超临界 主汽温度 鲁棒性 |
Application of a Neural Network Predictive Control in the Supercritical Main Steam |
| |
| LI Yun-Juan,FANG Yan-jun.Application of a Neural Network Predictive Control in the Supercritical Main Steam[J].Journal of Kunming Teachers College,2010,32(3):97-99,103. |
| |
| Authors: | LI Yun-Juan FANG Yan-jun |
| |
| Institution: | 1.Department of Automation Control and Mechanical Engineering,Kunming University,Yunnan Kunming 650118,China;2.Department of Automation,Wuhan University,Hubei Wuhan 430072,China) |
| |
| Abstract: | The traditional PID control is difficult in the non-linear,delay,time-varying and it has a disturbance characteristics of supercritical main steam temperature control system to achieve satisfactory control effect.So presents a neural network predictive control scheme,analysis of the algorithm theory and design process and the program of a multi-step prediction,rolling optimization and feedback correction control strategy,the reality of the robustness,good accuracy and fast control effect.Taking a supercritical main steam temperature as the research object,MATLAB simulation results show that： split the different conditions established by the main steam temperature of neural network dynamic model that can predict very well the dynamic characteristics of the object,and achieved better control effect than that of traditional PID. |
| |
| Keywords: | neural network predictive control supercritical fluid main steam temperature robustness |
| 本文献已被 维普 万方数据 等数据库收录！ |
|
