Condition Based Monitoring using Nonparametric Similarity Based Modeling


英字タイトル:
Condition Based Monitoring using Nonparametric Similarity Based Modeling
著者:
Stephan WEGERICH
発刊日:
公開日:
カテゴリ: 第3回
キーワードタグ:

概要

Nonparametric modeling techniques are utilized extensively in applications involving the analysis of complex data sets. Nonparametric approaches have the desirable characteristic that they do not require a priori knowledge of the analytical form of the model meant to charac terize a dataset. Instead, they rely exclusively on the available data to establish relationships present in the data. In the context of plantwide condition based monitoring, nonparametric modeling is an extremely powerful approach. Most, if not all systems within a plant th at are suitably instrumented and for which historical data exist, can be modeled and hence monitored. Similarity Based Modeling (SBM) is a particularly effective nonparametric condition monitoring technique. Unlike most nonparametric modeling approaches, SBM does not requir e complicated optimization algorithms to be trained and is routinely used to model systems with large numbers of variables. In this paper, the mathematics behind SBM is described and a performance comparison is presented between SBM and other condition monitoring approaches.


全文掲載のPDFファイルダウンロード