逐次ファジィ・ニューラルネットワークによるポンプの精密異常診断


英字タイトル:
Precise Diagnosis Method for Centrifugal Pump by Sequential Fuzzy Neural Network
著者:
神 豊 Yutaka JIN 薛 紅涛 Hongtao XUE 李 可 Ke LI 陳山 鵬 Ho JINYAMA 山村 尚広 Naohiro YAMAMURA 水津 明日香 Asuka SUIZU
発刊日:
公開日:
カテゴリ: 第10回
キーワードタグ:

概要

Abstract : This paper presents a precise condition diagnosis method for a centrifugal pump using sequential fuzzy neural network (SFNN). Firstly, the feature signals of fault states are extracted from the measured vibration signals for the pump condition diagnosis in different frequency regions by statistic filter, symptom parameters (SPs) for distinguishing fault types are defined and calculated by using the feature signals. Secondly, to distinguish the pump state, sequential fuzzy neural network (SFNN) is used to detect faults and identify fault types. The proposed method has been applied to detect the faults of a centrifugal pump, and the efficiency of the method has been verified using practical examples.


全文掲載のPDFファイルダウンロード