| 引用本文: | 韩敏, 韩冰.一种通用学习网络自适应算法及其在预测控制中的应用[J].控制理论与应用,2006,23(6):900~906.[点击复制] |
| HAN Min, HAN Bing.Adaptive algorithm of universal learning network and its application to predictive control[J].Control Theory and Technology,2006,23(6):900~906.[点击复制] |
|
| 一种通用学习网络自适应算法及其在预测控制中的应用 |
| Adaptive algorithm of universal learning network and its application to predictive control |
| 摘要点击 1590 全文点击 1066 投稿时间：2005-01-31 修订日期：2006-01-06 |
| 查看全文 查看/发表评论 下载PDF阅读器 |
| DOI编号 10.7641/j.issn.1000-8152.2006.6.010 |
| 2006,23(6):900-906 |
| 中文关键词 系统辨识 通用学习网络 大滞后 预测控制 |
| 英文关键词 system identification universal learning network long time delay predictive control |
| 基金项目 国家自然科学基金资助项目(60374064, 60674073). |
|
| 中文摘要 |
| 针对黑箱过程的辨识与控制, 本文提出了一种选择通用学习网络(universal learning network, ULN)节点间延迟时间参数的自适应算法, 并将其应用于对控制对象中的纯滞后参数的辨识. 将通用学习网络与PID控制器相结合, 应用于包含大滞后的系统的模型预测控制(model predictive control, MPC)中. 仿真结果证明通用学习网络能够有效地辨识被控对象的纯滞后时间, 并能够作为预估器应用于模型预测控制系统中. |
| 英文摘要 |
| An adaptive algorithm of universal learning network (ULN) is presented for time-delay parameter setting in this paper. Firstly, the ULN with PID controller is used in model predictive control (MPC) for stabilizing a class of nonlinear systems with long time-delay. Its application in identifying pure time-delay of the black box plant is then studied. Finally, simulation results show that the adaptive algorithm of the ULN model can identify the time-delay of the object model and be used as a predictor in predictive control system. |