Incorporating Known Plant Characteristics into Artificial Neural Networks for Dynamic System Identification and Control With Electrically Commutated Machine Applications

将已知的植物特征纳入人工神经网络，通过电换向机器应用进行动态系统识别和控制

• 批准号: 9216462
• 负责人: Ronald Brown
• 金额: $ 20万
• 依托单位: Marquette University
• 依托单位国家: 美国
• 项目类别: Continuing Grant
• 财政年份: 1993
• 资助国家: 美国
• 起止时间: 1993-02-15 至 1997-01-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=9216462&HistoricalAwards=false
• 关键词: Incorporating; Known; Plant; Characteristics; into

The primary emphasis of this project is on the use of artificial neural networks (ANNs) for identification and control of unknown nonlinear dynamic systems. However, generally much is known about the plant. This research involves the use of a "gray layer" which facilitates the integration of known characteristics of the otherwise unknown plant into an ANN. The convergence characteristics using this novel technique are of several orders of magnitude better than the conventional schemes. Both state of the art intelligent control and adaptive control technologies are used, and their application is demonstrated in the context of electrically commutated machines.*** //

该项目的主要重点是使用人工 神经网络（ANN）用于识别和控制未知 非线性动力系统然而，一般来说， 厂区这项研究涉及使用“灰色层”， 有助于集成已知的特性， 否则未知的植物到一个人工神经网络。收敛 使用这种新技术的特性是几个数量级 比传统的方案更好。两国 现有技术的智能控制和自适应控制技术 使用，并在上下文中演示了其应用程序 电动换向电机 * //