Non-intrusive Energy Disaggregation using Signal Unmixing Undergraduate: Philip Wolfe Mentors: Alireza Rahimpour, Yang Song Professor: Dr. Hairong Qi Final.

Slides:



Advertisements
Similar presentations
Machine Learning & Bioinformatics Tien-Hao Chang (Darby Chang) Machine Learning & Bioinformatics 1.
Advertisements

VORTEX Version Software Application Sociology; Marketing research; Social-psychological research Social-medical research Staff recruitment, staff.
Context-based object-class recognition and retrieval by generalized correlograms by J. Amores, N. Sebe and P. Radeva Discussion led by Qi An Duke University.
Aggregating local image descriptors into compact codes
DIMENSIONALITY REDUCTION: FEATURE EXTRACTION & FEATURE SELECTION Principle Component Analysis.
Data Mining Methodology 1. Why have a Methodology  Don’t want to learn things that aren’t true May not represent any underlying reality ○ Spurious correlation.
Content-based retrieval of audio Francois Thibault MUMT 614B McGill University.
Outline Introduction Music Information Retrieval Classification Process Steps Pitch Histograms Multiple Pitch Detection Algorithm Musical Genre Classification.
Introduction The aim the project is to analyse non real time EEG (Electroencephalogram) signal using different mathematical models in Matlab to predict.
SLIQ: A Fast Scalable Classifier for Data Mining Manish Mehta, Rakesh Agrawal, Jorma Rissanen Presentation by: Vladan Radosavljevic.
1 Framework for Energy Data Collection Prashant Lodhia Rebekah Drake.
Raster Data. The Raster Data Model The Raster Data Model is used to model spatial phenomena that vary continuously over a surface and that do not have.
Supervised classification performance (prediction) assessment Dr. Huiru Zheng Dr. Franscisco Azuaje School of Computing and Mathematics Faculty of Engineering.
Retrieval Evaluation. Brief Review Evaluation of implementations in computer science often is in terms of time and space complexity. With large document.
Simultaneous Localization and Map Building System for Prototype Mars Rover CECS 398 Capstone Design I October 24, 2001.
Results The following results are for a specific DUT device called Single Ring Micro Resonator: Figure 6 – PDL against Wavelength Plot Figure 7 – T max.
A Hadoop MapReduce Performance Prediction Method
What is machine learning? 1. A very trivial machine learning tool K-Nearest-Neighbors (KNN) The predicted class of the query sample depends on the voting.
Design and Development of an Accelerometer based Personal Trainer System By Emer Bussmann B.E. Electronic Engineering April 2008.
Graphical Tree-Based Scientific Calculator: CalcuWiz Will Ryan Christian Braunlich.
Host Load Prediction in a Google Compute Cloud with a Bayesian Model Sheng Di 1, Derrick Kondo 1, Walfredo Cirne 2 1 INRIA 2 Google.
CS 5604 Spring 2015 Classification Xuewen Cui Rongrong Tao Ruide Zhang May 5th, 2015.
Identifying Computer Graphics Using HSV Model And Statistical Moments Of Characteristic Functions Xiao Cai, Yuewen Wang.
Prognostics of Aircraft Bleed Valves Using a SVM Classification Algorithm Renato de Pádua Moreira Cairo L. Nascimento Jr. Instituto Tecnológico de Aeronáutica.
An Adaptive Modeling for Robust Prognostics on a Reconfigurable Platform Behrad Bagheri Linxia Liao.
Performance Evaluation of Hybrid MPI/OpenMP Implementation of a Lattice Boltzmann Application on Multicore Systems Department of Computer Science and Engineering,
EnSight analyze, visualize, communicate EnSight 6.x Advanced Training Part 1 Instructors: Mike Krogh, Anders Grimsrud.
Network Intrusion Detection Using Random Forests Jiong Zhang Mohammad Zulkernine School of Computing Queen's University Kingston, Ontario, Canada.
Performance and Insights on File Formats – 2.0 Luca Menichetti, Vag Motesnitsalis.
 The most intelligent device - “Human Brain”.  The machine that revolutionized the whole world – “computer”.  Inefficiencies of the computer has lead.
Hand Tracking for Virtual Object Manipulation
Permission-based Malware Detection in Android Devices REU fellow: Nadeen Saleh 1, Faculty mentor: Dr. Wenjia Li 2 Affiliation: 1. Florida Atlantic University,
Treatment Learning: Implementation and Application Ying Hu Electrical & Computer Engineering University of British Columbia.
Feature selection LING 572 Fei Xia Week 4: 1/29/08 1.
Computational Model of Energetic Particle Fluxes in the Magnetosphere Computer Systems Yu (Evans) Xiang Mentor: Dr. John Guillory, George Mason.
A hybrid SOFM-SVR with a filter-based feature selection for stock market forecasting Huang, C. L. & Tsai, C. Y. Expert Systems with Applications 2008.
Tone Mapping on GPUs Cliff Woolley University of Virginia Slides courtesy Nolan Goodnight.
Combining multiple learners Usman Roshan. Bagging Randomly sample training data Determine classifier C i on sampled data Goto step 1 and repeat m times.
Dana Butnariu Princeton University EDGE Lab June – September 2011 OPTIMAL SLEEPING IN DATACENTERS Joint work with Professor Mung Chiang, Ioannis Kamitsos,
Sentosa Technology Consultants | | KDDI R&D Laboratories Inc. Automatic Content Filtering KDDI R&D Laboratories Inc.
Enrique Martínez-Meyer
Logic Analyzer ECE-4220 Real-Time Embedded Systems Final Project Dallas Fletchall.
ISCG8025 Machine Learning for Intelligent Data and Information Processing Week 3 Practical Notes Application Advice *Courtesy of Associate Professor Andrew.
Project by: Cirill Aizenberg, Dima Altshuler Supervisor: Erez Berkovich.
BOĞAZİÇİ UNIVERSITY DEPARTMENT OF MANAGEMENT INFORMATION SYSTEMS MATLAB AS A DATA MINING ENVIRONMENT.
A Semi-Blind Technique for MIMO Channel Matrix Estimation Aditya Jagannatham and Bhaskar D. Rao The proposed algorithm performs well compared to its training.
A NOVEL METHOD FOR COLOR FACE RECOGNITION USING KNN CLASSIFIER
Dutch Protocol Monitoring Energy Savings & EU-directive P. Boonekamp ECN, Netherlands EU/ECEEE-seminar , Brussels.
GENDER AND AGE RECOGNITION FOR VIDEO ANALYTICS SOLUTION PRESENTED BY: SUBHASH REDDY JOLAPURAM.
ApproxHadoop Bringing Approximations to MapReduce Frameworks
Learning Photographic Global Tonal Adjustment with a Database of Input / Output Image Pairs.
Demand Response Analysis and Control System (DRACS)
Energy Consumption Forecast Using JMP® Pro 11 Time Series Analysis
Reza Yazdani Albert Segura José-María Arnau Antonio González
CEN352 Dr. Nassim Ammour King Saud University
Bag-of-Visual-Words Based Feature Extraction
Date of download: 12/22/2017 Copyright © ASME. All rights reserved.
Fabien LOTTE, Cuntai GUAN Brain-Computer Interfaces laboratory
Dejavu:An accurate Energy-Efficient Outdoor Localization System
Multidisciplinary Engineering Senior Design Project P06441 See Through Fog Imaging Preliminary Design Review 05/19/06 Project Sponsor: Dr. Rao Team Members:
Basic machine learning background with Python scikit-learn
Students: Meiling He Advisor: Prof. Brain Armstrong
הודעות ריענון מהיר והרחבות דגימת אות Low-Level
Tze Meng Low, Qi Guo, Franz Franchetti
Xin Qi, Matthew Keally, Gang Zhou, Yantao Li, Zhen Ren
MAS 622J Course Project Classification of Affective States - GP Semi-Supervised Learning, SVM and kNN Hyungil Ahn
Analysis on Accelerated Learning Cohorts
Using Clustering to Make Prediction Intervals For Neural Networks
Random Neural Network Texture Model
MyoHMI Architecture Background
Presentation transcript:

Non-intrusive Energy Disaggregation using Signal Unmixing Undergraduate: Philip Wolfe Mentors: Alireza Rahimpour, Yang Song Professor: Dr. Hairong Qi Final Presentation- 7/17/14

Project Description

Why? ◦Increase consumer knowledge ◦Effect change in consumer behavior ◦Save energy ◦Gather valuable data

Data Management and Visualization ◦Given ◦REDD- public dataset for disaggregation research ◦Problems ◦UTC format, unevenly spaced intervals (1, 3, 4 seconds, geographic location ◦house data ~1.5e6, memory management, increased calculation time ◦Evaluating results and data inefficient ◦Solutions ◦Convert to serial time, and time zone- Boston (EST) ◦Vectorization, downsample (1/60 Hz)

Data Management and Visualization ◦Implement GUI to run through indexes and grab desired data ◦Channel mains and device indexing mismatch ◦Day prediction algorithm ◦Variable number channel input handling ◦Visualization GUI to plot each device ◦Checkbox callback channel plotting ◦Timescale axis conversion ◦Color dictionary selection/cycling and Legend ◦Variable number channel input handling

Approach

Supervised Training

Generalization

Results Figure 1. Aggregate House 6, Day 1.

Results Figure 2. Individual Device Comparison. House 6, Day 1.

Results Figure 3. Estimate vs True device consumption

Conclusion

Future Work ◦Optimize computational time ◦Multicore processing ◦Find alternative feature extraction method ◦More powerful computer (currently 2.4GHz) ◦Increase accuracy ◦Use higher sampling rate for calculations (downside computation time) ◦Additional data ◦Manually identify key patterns/features (not optimal) ◦Create specialized signal feature extraction for each device

Questions