Automated Classification and Analysis of Power System Disturbance Waveforms Using a Neural Network Approach

使用神经网络方法对电力系统扰动波形进行自动分类和分析

• 批准号: 9412764
• 负责人: David Lubkeman
• 金额: $ 8.74万
• 依托单位: Clemson University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 1994
• 资助国家: 美国
• 起止时间: 1994-09-15 至 1998-05-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=9412764&HistoricalAwards=false
• 关键词: Automated; Classification; Analysis; Power; System

9412764 Lubkeman The objective of this proposed project is to investigate and develop a scheme to further automate the monitoring, recording, and diagnosis of power disturbances for power quality studies, using a combination of artificial neural network (ANN) and case- based reasoning (CBR) approaches. The proposed research is (a) to develop a scheme to automate the data collection process of recorders--this involves an ANN methodology for the classification of waveforms that are captured and a rule-based relational database design for the effective archival of data, and (b) to investigate the applicability of a case-base reasoning approach towards aiding the diagnosis of disturbances. This research represents a step in the direction of intelligent diagnostic and measurement tools using state-of- the-art artificial intelligence technologies which are just emerging. It would be a significant improvement over present-day recorders, providing for further automation of the measurement and diagnosis of disturbances. ***

9412764 Lubkeman该项目的目的是研究和开发一种方案，以进一步自动化的监测，记录和诊断电力扰动的电能质量研究，使用人工神经网络（ANN）和基于案例的推理（CBR）的方法相结合。 拟议的研究是（a）开发一个计划，以自动化的数据收集过程的记录-这涉及到一个人工神经网络的方法被捕获的波形和基于规则的关系数据库设计的有效存档的数据的分类，和（B）调查的适用性的情况下，基于推理的方法，以帮助诊断的干扰。 这项研究代表了智能诊断和测量工具的方向迈出了一步，使用了刚刚出现的最先进的人工智能技术。 这将是对目前记录器的重大改进，为进一步自动化测量和诊断扰动提供了条件。 ***