Please use this identifier to cite or link to this item: http://hdl.handle.net/123456789/11198
Title: CONDITION MONITORING AND FAULT DIAGNOSES OF ROTATING MACHINERY OF A THERMAL POWER PLANT USING MODEL-BASED ORDER TRACKING SYSTEM
Authors: Kumar, Manish
Keywords: MECHANICAL INDUSTRIAL ENGINEERING
ROTATING MACHINERY
THERMAL POWER PLANT
MODEL-BASED ORDER TRACKING SYSTEM
Issue Date: 2005
Abstract: Due to presence of competitive environment, high productivity and quality production have become necessary requirements of the industries today. Proper and timely maintenance of the machines is imperative to achieve this goal. Early fault detection in a machine is a key element of effective maintenance and the diagnosis of the machine faults at an early stage has become necessary requirement of maintenance strategy. There are many standard fault diagnosis techniques available today._ Vibration Signature Analysis is one of them. The present work is based on this technique. This work focuses on the development of a model-based order tracking system. The system is composed of a signal processing module and a state inference module. In the signal processing module, the recursive least square algorithm is used to extract the amplitude of vibration signals. This method is verified experimentally to diagnose the fault of rotating equipments such as induced draught fan, primary air fan, circulating water pump, drip pump, ash slurry pump, and scanner air fan in a thermal power plant. The vibration data i.e. horizontal, vertical and axial displacement and acceleration, phase angle and frequency spectrum has been taken by accelerometer and Vibroport 41. According to the technical specification of the above equipment, collected data has been compared and checked with ISO 2372 chart. Further, signature analysis has been done for predicting the type of existing problems. In the iv vibration signature analysis, cepstrum analysis has been done for simplifying the complex vibration signature and detecting the harmonics of exciting .frequency and amplitude of each peak with its relationship to the fundamental existing frequency. The complex vibration signatures are processed in MATLAB environment to explore valuable information by simplifying the vibration signal. Results obtained are as per the expectation. An attempt has been made with the above model based tracking system to detect the " various types of vibration problems such as unbalance, misalignment of coupling, mechanical looseness of components and foundation, bearing problems and bent shaft. A detailed analysis has been carried out with the help of above data and attempt has been made to establish the particular type of problem with its characteristic frequency
URI: http://hdl.handle.net/123456789/11198
Other Identifiers: M.Tech
Appears in Collections:MASTERS' DISSERTATIONS (MIED)

Files in This Item:
File Description SizeFormat 
MIEDG12431.pdf5.79 MBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.