Masters Theses

Abstract

"This work presents an appliance disaggregation technique to handle the fundamental goal of the Non-Intrusive Appliance Load Monitoring (NIALM) problem i.e., a simple breakdown of an appliance level energy consumption of a house. It also presents the modeling of individual appliances as load models using hidden Markov models and combined appliances as a single load model using factorial hidden Markov models. Granularity of the power readings of the disaggregated appliances matches with that of the readings collected at the service entrance. Accuracy of the proposed algorithm is evaluated using publicly released Tracebase data sets and UK-DALE data sets at various sampling intervals. The proposed algorithm achieved a success rate of 95% and above with Tracebase data sets at 5 second sampling resolution and 85% and above with UK-DALE data sets at 6 second sampling resolution"--Abstract, page iii.

Advisor(s)

McMillin, Bruce M.

Committee Member(s)

Kimball, Jonathan W.
Chellappan, Sriram

Department(s)

Computer Science

Degree Name

M.S. in Computer Science

Publisher

Missouri University of Science and Technology

Publication Date

Spring 2015

Pagination

vii, 34 pages

Note about bibliography

Includes bibliographical references (pages 32-33).

Rights

© 2015 Anusha Sankara, All rights reserved.

Document Type

Thesis - Open Access

File Type

text

Language

English

Subject Headings

Household appliances -- Energy consumption -- Computer simulation
Hidden Markov models
Energy consumption -- Measurement

Thesis Number

T 10692

Electronic OCLC #

913514687

Share

 
COinS