Use this URL to cite or link to this record in EThOS: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.422359
Title: Aerodynamic investigation of fluid injection in an axial compressor
Author: Mustafa, Zainol
Awarding Body: Cranfield University
Current Institution: Cranfield University
Date of Award: 2005
Availability of Full Text:
Access through EThOS:
Access through Institution:
Abstract:
The main objective of the investigation is to provide a physically realistic model describing the movement of liquid phase in an industrial axial compressor during online washing. ln order to achieve this objective, CFD simulations were developed for predicting the water particle trajectories inside an industrial axial compressor, the rate of evaporation of water droplets on axial compressor blades by inertial impaction, turbulent diffusion, pressure and temperature increments. Simulation of water droplet trajectories and evaporation of water droplet content in designed 260MW gas turbine engine were undertaken. Two types of boundary conditions have been considered for a droplet. The first is the inlet boundary condition, which describes the properties of a droplet at the entry/inlet position. It was assumed that the droplets are evenly distributed at the inlet face. For the droplet inlet boundary, there were 4 main factors considered, namely droplet initial temperature, droplet flow rate, droplet axial slip velocity and droplet average initial diameter. Each factor has at least two levels, namely the base line level and some deviation or deviations to the base-line level. The second boundary condition is the wall boundary condition, which models what happens when a droplet interacts with a solid wall. For this second type boundary condition, all simulations in this study, the coefficient of restitution of all solid walls was set at very low number, coefficient of restitution = 0.0005. It is assumed that a droplet that hits a solid wall will undergo a non-perfect rebound. Other than mechanical effects due to the impact of particulates during online washing, the main aero thermal effects are due to the flow changes occurring on account of the two phase nature of the working fluid, heat and mass transfer among the two phases, modification in thermal and transport properties, and changes in chemical action and combustion. Practically, it is not possible, in general, to assure that all of the washing fluid, which entered the engine during online washing, would always be converted to the gas phase by the time the working fluid leaves the high pressure compressor, or even the combustor. Thus, both the state of the fluid and the cross-sectional distribution of washing fluid are of concern throughout the engine flow path, especially the first 3 front stages of the axial compressor. It must be noted that the preceding statements take no account of the presence and effects of bypass doors, vents, and valves. This is a set of devices for which there are no reliable data for air-fluid mixture operation, and thus, no guidance in design or installation. They can introduce major changes in the amount and distribution of the discrete phases of fluid along the flow path in compressors. Finally, however there is a need for at least one set of definitive tests on selected components and engines. At the moment, there is neither clear view nor experimental data available to validate those said objectives. On one hand, there is no validated computational model available to assist with understanding the complex relationship between the injection parameters and the resultant flow pattern and trajectories of the droplet as well as there is no information regarding the droplet build up on the blade surface in a multistage axial compressor during online washing. Nevertheless, even though without experimental validation, this investigation is very important as advancing tools to further understand the axial compressor online washing phenomenon which is still not fully understood even today. Therefore, post-axial-compressor fluid motion is important in understanding the role of air/fluid mixing and fluid evaporations and will be analysed and presented in this study. More precisely, this investigation tries to quantify the effect of injected droplet parameters on generic multistage axial compressor. As this work is focused on the model (droplet parameter) sensitivities rather than the CFD model itself, presented in this investigation herewith are only the main features of the generic droplet trajectories flow pattern in a multistage industrial axial compressor.
Supervisor: Pilidis, P. Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.422359  DOI: Not available
Share: