Please use this identifier to cite or link to this item:
Signal Processing and Diagnosis Simulation od a Machine System (III)
|Contributors: ||NTOU:Department of Systems Engineering and Naval Architecture|
|Keywords: ||Auto-Regressive model with eXogenous model (ARX);Self Organizing Map (SOM);Pumping system;Identification;Diagnosis|
|Issue Date: ||2011-06-28T08:19:40Z
|Abstract: ||In this three-years research project, inherent characteristic identification and fault diagnosis of a scale-modeled marine pumping system have been accomplished. Signals selected for the processing of identification and diagnosis of the set-up pumping system driven by a motor encompass the measured data of voltage, current, vacuum pressure and flow rate. An Auto-Regressive model with eXogenous (ARX) is used to approximate the dynamic behavior of the system. Owing to the states or the parameters of a monitoring process are close to the system fault of signal flow, thus the observer-based state and system's parameters estimation theories are well developed in Fault Detection and Isolation (FDI) for Linear Time Invarient (LTI) systems. However the systems, where explicit models are difficult to derive, cannot be directly implemented by the redundancy based FDI strategy of a LTI system. In the aspects of diagnosis, classified weight vectors of Self Organizing Map (SOM) and a neural network approach are presented to diagnose the performance of the pumping system in relation to the system parameters, i.e., pressure, vacuum, flow rate, voltage and current, etc, with or without faults in the system ,The merits of this approach through SOM is that the information of the pumping system condition is embedded in the relations amongst system variables. One can easily diagnose the fault of the pumping system via the clearly classified weight vectors form the SOM. Furthermore, the correlation between the outcomes of system identification and the measured signals are coincident well in the output model. Also, this established approach can be used to diagnose the performance of the pumping system with malfunction without experience.|
|Appears in Collections:||[系統工程暨造船學系] 研究計畫|
Files in This Item:
There are no files associated with this item.
All items in NTOUR are protected by copyright, with all rights reserved.