No Cover Image

E-Thesis 843 views 411 downloads

Health monitoring techniques for rotating machinery. / Jyoti K Sinha

Swansea University Author: Jyoti K Sinha

Abstract

The present research is concerned with health monitoring techniques for rotating machinery, for example Turbogenerator (TG) sets in the power industry. Vibration based condition monitoring is widely accepted for rotating machinery and hence the vibration response of a machine is again utilized in th...

Full description

Published: 2002
Institution: Swansea University
Degree level: Doctoral
Degree name: Ph.D
URI: https://cronfa.swan.ac.uk/Record/cronfa42376
Abstract: The present research is concerned with health monitoring techniques for rotating machinery, for example Turbogenerator (TG) sets in the power industry. Vibration based condition monitoring is widely accepted for rotating machinery and hence the vibration response of a machine is again utilized in the present research study. Experience shows that faults develop in rotating machines during normal operation and hence their quick identification and remedy are important from safety and plant productivity considerations. The vibration based fault identification procedures are well developed for rotating machinery. However the quantification part of the identified faults has still not matured, and is an ongoing research topic. Hence the remedial action is usually time consuming, even though the machine is known to have some known faults, due to lack of knowledge of their locations and the extent of the faults. In general such a quantification of the identified fault relies on the mathematical model of the complete system along with the measured vibration response of the system. Rotating machinery consists of three major parts - a rotor, fluid journal bearings and a foundation which is often flexible. Often a good model of the rotor (usually a finite element model) and an adequate model of the fluid bearings may be constructed. However, a reliable model for the foundation is difficult to construct due to a number of practical difficulties. Hence the present study has concentrated on two objectives - reliable modelling for the foundation and the quantification of faults using the measured vibration response at the bearing pedestals and the mathematical model of the rotor and the fluid bearings. For the foundation model, the theory which was developed to estimate the models for flexible foundation has been described in the thesis. The method uses measured vibration response at bearing pedestals during machine run-downs, a priori rotor and journal bearing models, and a knowledge of the rotor unbalance, to estimate the stiffness, damping and mass matrices of the foundation. The method was tested on both simulated and experimental examples. The prediction capability of the estimated foundation model was also demonstrated. For the fault estimation a different approach has been used. It has been assumed that the foundation mathematical model is not known, and it is demonstrated that the two faults - the state of rotor unbalance and the misalignment in the rotor can be estimated reliably. The theory of the proposed methods is discussed in the thesis. The method uses measured vibration response at bearing pedestals during a single machine run-down, and a priori rotor and journal bearing models, to estimate the rotor unbalance and the misalignment along with the foundation parameters, so that the dynamics of the foundation is also accounted for during the estimation. The methods were tested on simulated and experimental examples and the estimation accuracy was found to be excellent and generally robust to errors in the rotor and bearing models.
Keywords: Mechanical engineering.
College: Faculty of Science and Engineering

Similar Items