Determining Airfoil Self-Noise Levels

Slides:



Advertisements
Similar presentations
University of Western Ontario
Advertisements

Michael DeRosa Master of Engineering Final Project Exploration of Airfoil Sections to Determine the Optimal Airfoil for Remote Controlled Pylon Racing.
Kyle Desrosiers (ME)- Team Lead Brad Fiedler (EE) Greg Wodzicki (EE) Chris VanWagenen (EE) George Slack- Faculty Guide P ACTIVE NOISE CANCELLATION.
x – independent variable (input)
Regionalized Variables take on values according to spatial location. Given: Where: A “structural” coarse scale forcing or trend A random” Local spatial.
Face Recognition Using Neural Networks Presented By: Hadis Mohseni Leila Taghavi Atefeh Mirsafian.
© 2011 Autodesk Freely licensed for use by educational institutions. Reuse and changes require a note indicating that content has been modified from the.
Waves & Wave Properties Determining the Speed of Sound Lab
CFD Pre-Lab 2 Simulation of Turbulent Flow around an Airfoil Seong Mo Yeon, and Timur Dogan 11/12/2013.
Introduction to Fluid Mechanics
Motion.
Development of a Baseline Tropical Cyclone Model Using the Alopex Algorithm Robert DeMaria.
VIBRATIONS EXPERIMENT OBJECTIVES: 1. Solve a second order non-homogenous differential equation describing the displacement of a specimen.
Measurement of Pressure Distribution and Lift for an Airfoil  Purpose  Test design  Measurement system and Procedures  Uncertainty Analysis  Data.
CS121 Quiz 3 Fall 2012 Quiz Tips. Quiz 1 Tips Question 1 From the many calculation choices, use eval to calculate the dependent variable P(s), but use.
CS 121 – Quiz Feb 17 th. Question 2 We want to count how many days there were such that the temperature is at least degrees lower than the.
PHYSICS CLASS ACTIVITY. CLASS ACTIVITY: TUNING FORK FREQUENCY.
Two Least Squares Applications Data Fitting Noise Suppression.
Chapter 11 – Neural Nets © Galit Shmueli and Peter Bruce 2010 Data Mining for Business Intelligence Shmueli, Patel & Bruce.
Kinematics Vocabulary Units and vectors and scalars, oh my!
Airfoil in a Wind Tunnel Experiment #6
Airfoil in a Wind Tunnel
IIS for Speech Processing Michael J. Watts
Combining Models Foundations of Algorithms and Machine Learning (CS60020), IIT KGP, 2017: Indrajit Bhattacharya.
Date of download: 10/1/2017 Copyright © ASME. All rights reserved.
Regression and Correlation of Data Summary
SUR-2250 Error Theory.
Chapter 7. Classification and Prediction
33. Conic Sections in General Form
Probability and Statistics for Computer Scientists Second Edition, By: Michael Baron Section 11.1: Least squares estimation CIS Computational.
Lecture 19. Review CS 109A/AC 209A/STAT 121A Data Science: Harvard University Fall 2016 Instructors: P. Protopapas, K. Rader, W. Pan.
Cart on Ramp Lab.
Motion Graphs Position-Time (also called Distance-Time or Displacement-Time) d t At rest.
Graphical Analysis Of Motion
CSE 4705 Artificial Intelligence
Analyzing Redistribution Matrix with Wavelet
Basic machine learning background with Python scikit-learn
Analyzing the Sound Pressure of Airfoil Self-Generated Noise
Bias and Variance of the Estimator
Machine Learning – Regression David Fenyő
Probability and Statistics for Computer Scientists Second Edition, By: Michael Baron Section 11.1: Least squares estimation CIS Computational.
AIRFRAME NOISE MODELING APPROPRIATE FOR MULTIDISCIPLINARY DESIGN AND OPTIMIZATION AIAA Serhat Hosder, Joseph A. Schetz, Bernard Grossman and.
Simulation of turbulent airfoil Flow Using FlowLab 1.1 (CFD PreLab 2)
Lecture 10: Observers and Kalman Filters
Statistical Methods For Engineers
Foundations of Physical Science
Lecture 10 2D plotting & curve fitting
FIDAP Numerical Modeling
Kinematics Acceleration
Motion.
Sampling Distribution
Sampling Distribution
Department of Computer Science University of York
Motion.
Simple Harmonic Motion
Physics Intro & Kinematics
Neural Networks Geoff Hulten.
Ch. 11: Motion 11.1: Distance and Displacement
Let’s Graph Unit 1.
Overfitting and Underfitting
2. KINEMATICS AND KINETICS
CHAPTER-16 T071,T062, T061.
Upscaling Petrophysical Properties to the Seismic Scale
What does the word “acceleration” mean to you?
Section 8, Lecture 1, Supplemental Effect of Pressure Gradients on Boundary layer • Not in Anderson.
MECH 373 Instrumentation and Measurement
Real-time Uncertainty Output for MBES Systems
Position, Speed and Velocity
Motion in One Dimension
Probabilistic Surrogate Models
Presentation transcript:

Determining Airfoil Self-Noise Levels Josh Merchant

Executive Summary The goal of the project is to develop a better way to predict the level of noise that is produced by an airfoil in high velocity air Completed – data preprocessing, model development, general testing and tuning of the model To Do – fine tuning of the model, in depth performance analysis Model is performing as well or better than the target of 0.89 Coefficient of Determination

Data Data was taken from a NASA experiment to determine self-noise levels Data needed to be scaled and centered to allow for learning from all variables Order of the data was randomized Original test run to minimize setup changes   Frequency (Hz) Angle of Attack (Degrees) Chord Length (Meters) Free Stream Velocity (m/s) Suction Side Displacement Thickness (m) Sound Pressure (dB) Average 2886.38 6.78 0.14 50.86 0.01 124.84 Sd 3151.52 5.92 0.09 15.57 6.90 Scaled Sd Scaled 1

Method scikit learn MLPRegressor Used CS computers Written in Python Utilized numpy functions Used CS computers Small models, about 1000 samples 10 second run time without graphing Iteratively optimized parameters Made gross changes with scripts to see initial trends Fine tuning the model by hand Current best performance Activation Function – Logistic Nodes – 5 – 15 – 15 – 1 Solver Function – lbfgs Learning Rate – Constant Initial learning Rate – 0.001

Results Model has a Coefficient of Determination of 0.95 Model has a Mean Squared Error of 0.05 Performing better than target network which had coefficient of Determination of 0.89 Mean Squared Error of 0.08 12 or less nodes resulted in underfitting 25 or more nodes resulted in overfitting

Discussion Data processing is Important Tuning Parameters Future Non-Scaled and Non-Centered models had Negative Coefficient of Determination Non-Randomized models had Negative Coefficient of Determination Tuning Parameters Not one dominant parameter Different combinations have similar results Future Tweak Parameters In Depth Analysis