Analysis of the Dynamic Behavior of Two High-Pressure Turbines for the Possible Detection of Rub Symptoms

Abstract

Physical contact between the rotor and stationary-elements is a main topic in the field of rotordynamics. In the last decades, many studies have investigated the causes and the consequences of this phenomenon; the so-called ‘rub’, resulting in several theories and mathematical models. Muszynska characterized the rub from a phenomenological point of view in different conditions, both steady-state and transient; her studies laid the foundations and inspired research in the following years in many universities and research centers. State-of-the-Art methods permit to recognize, in a simple way, the rub during machine operating condition but in spite of the amount of studies, it is still difficult to prevent the occurrence of rubbing. This paper focuses on identifying what are the forerunner symptoms of rubbing without attach specific attention to the rub and even to the consequences on the machine behavior. The authors investigate how to prevent the occurrence of rub by detecting its symptoms by means of analysis of the vibration and of the diagnostic data. For this purpose, the conventional graphic tools are used: Bode plot, polar plot, cascade plot, waterfall plot, centerline plot, time-based plot etc. The dynamic behavior of two high-pressure steam turbines, which have been damaged by a physical contact between the rotor and a stationary part, is analyzed. In order to investigate the evolutionary trend of the vibration for possible rub detection, a new graphic tool was developed by the authors. This work is born thanks to an industrial interest about the identification of evidences in the dynamic behavior of the forthcoming rub. The data are available in three different sets of measurements, for both the high-pressure turbines. The data cover a quite long period but they are not available during the rubbing. It must be noticed that the rub has occurred, for both turbines, immediately after a trip of the machine, so it is not excludable, a priori, that the cause of the trip has not been the same cause of the rubbing.