An enhanced ISODATA algorithm for recognizing multiple electric appliances from the aggregated power consumption dataset

Mingchao Li, Shuai Han, Jonathan Shi

Research output: Journal article publicationJournal articleAcademic researchpeer-review

23 Citations (Scopus)

Abstract

An electric meter is used for measuring the electricity consumption in a household. The obtained total data doesn't separate the consumptions of individual appliances and provides little actionable information for energy efficiency improvement. This paper presents an enhanced Iterative Self-Organizing Data Analysis (ISODATA) algorithm which can identify the operating modes and electricity consumptions of major appliances based on the metered total electricity consumption. The enhanced ISODATA algorithm is improved with k-medoids and parameter selections. It can overcome various drawbacks of the original k-means and the original ISODATA algorithm. Validated with actual data, the presented method is able to disaggregate the total electricity consumption efficiently and accurately into categories corresponding to major end uses.

Original languageEnglish
Pages (from-to)305-316
Number of pages12
JournalEnergy and Buildings
Volume140
DOIs
Publication statusPublished - 1 Apr 2017

Keywords

  • Clustering
  • Electrical appliance
  • ISODATA
  • Pattern recognition
  • Unsupervised learning

ASJC Scopus subject areas

  • Civil and Structural Engineering
  • Building and Construction
  • Mechanical Engineering
  • Electrical and Electronic Engineering

Cite this