ORCA Proposal
Developing Finite Element Meshes for the Acoustic Analysis of Jet Engines
Submitted by: Karl Kowallis
Modern Jet engines have been around since the 1930's and little has changed. The life span of a well-designed engine can be several decades and most engines in use today can trace their origin to the 1970's. The methods available for analysis in the early history of aircraft engine design included analytical calculations and modeling. They made many simplifying assumptions in the geometry of analytical models. The development of computer design applications allow a bridge in development by providing the computing power to accurately create and model jet engines.
Definition of Problem
By modeling a General Electric CF34 Turbofan, I will participate in a research project to analyze the acoustic modes of the engine, and determine what alterations to the design will improve performance. Using CAD software I will develop a FEM(Finite Element Mesh) model that we can study in a numerical acoustics package called SYSNOISE.
My development is part of a larger project of Dr. Scott Sommerfeldt, of the Brigham Young University Physics Department. It is also a beginning for the dissertation work of Lauralee Ireland. Due to the breadth of the problem I will only be able to complete the mesh development and assist in the initial analysis. The software we are using is new to the Physics Department and I have spent several months learning how to use it to benefit the project. My objective will be to finish the models so they can be used by the other project members.
Significance of Problem
Most models use a simple cylinder to approximate the engine including D.L. Sutliff's recent analysis of active control of a low speed turbofan(1). By modeling the actual geometry of an engine, we expect greater accuracy that may lead to changes having a significant effect on future engine design.
SYSNOISE and also I-DEAS, the package used to create the meshes, are powerful software packages. Not only will the development work on this aircraft engine be valuable, but the work will help us to learn these packages. Knowing how to use SYSNOISE and I-DEAS will provide greater options for solving future problems in acoustics.
Rationale of Proposed Solution
Using SDRC's software CAD package I-DEAS, I will create several meshes of the CF34 turbofan. Exporting these meshes in a standard file format will allow them to be imported by SYSNOISE. Once in SYSNOISE the appropriate boundary conditions can be entered to model the acoustics and air flow of the engine.
Expected Results
I anticipate developing approximately a dozen meshes of various nodal densities. Using the simplest model I will use SYSNOISE to analyze the lowest order mode. These initial calculations will allow assessment of my assumptions on boundary conditions so that when the more complex models are studied they will be more accurate.
SYSNOISE has extensive postprocessing capabilities to graph and chart acoustic variations. These graphs will provide an intuitive check to consider the validity of the results. Dr. Scott Sommerfeldt, Lauralee Ireland and I will discuss the results, their meaning, and significance.
If our results are repeatable and consistent we will be able to introduce a new method of analysis in the acoustic field. GE has provided us with our model data and is interested in the results.
My development will be included in my Senior Honors Thesis. My proposal to the Honors Department has been accepted and my thesis title is, __Developing Finite Element Meshes for the Acoustic Analysis of Jet Engines__. My work will be reported to the Physics department so that they can utilize my research for future reference.
Contingency Plans
My development on the meshes has already begun. If needed, parts of the airflow path that are being modeled will be simplified. We also have the option of modeling a cylindrical engine model that I initially created to practice using the CAD software. We could compare the results of that model with those obtained elsewhere. Dr. Sommerfeldt has indicated that learning how to use the software is one of the most important goals that will be achieved, regardless of our research results.
Resources Available
In preparation for this project the acoustics group UNIX server has been upgraded. Over the summer, while I was learning the CAD software, I worked as an assistant system administrator and installed a new operating system, the SYSNOISE software, and have begun to install I-DEAS. (We have been using the Engineering Departments software up to this point.) In order to make room for the software we also added a new hard drive to more than triple our available file space.
BYU has site licenses for both I-DEAS and SYSNOISE. SYSNOISE is installed and I-DEAS is in the process of being installed on our server. If we are unable to install I-DEAS, we may continue to use the Engineering Departments copy.
1. Sutliff, D.L. et al. __Active Noise Control of Low Speed Fan Rotor-Stator Modes__. 3rd AIAA/CEAS Aeroacoustics Conference, Atlanta, GA, 1997. AIAA-97-1641.
1999 Karl Kowallis <kowallis (a) mit edu>
All Rights Reserved
