Revealing Household Characteristics from Electricity Meter Data with Grade Analysis and Machine Learning Algorithms

Abstract

In this article, the Grade Correspondence Analysis (GCA) with posterior clustering and visualization is introduced and applied to extract important features to reveal households’ characteristics based on electricity usage data. The main goal of the analysis is to automatically extract, in a non-intrusive way, number of socio-economic household properties including family type, age of inhabitants, employment type, house type, and number of bedrooms. The knowledge of specific properties enables energy utilities to develop targeted energy conservation tariffs and to assure balanced operation management. In particular, classification of the households based on the electricity usage delivers value added information to allow accurate demand planning with the goal to enhance the overall efficiency of the network. The approach was evaluated by analyzing smart meter data collected from 4182 households in Ireland over a period of 1.5 years. The analysis outcome shows that revealing characteristics from smart meter data is feasible, and the proposed machine learning methods were yielding for an accuracy of approx. 90% and Area Under Receiver Operating Curve (AUC) of 0.82.

Simulation Study on the Electricity Data Streams Time Series Clustering

“…Our analysis showed that there are noticeable differences between consumers' behaviors, allowing us to distinguish homogeneous groups. In order to meet these challenges and to balance the system, customer profiling seems to be the solution to the problem of instability in power systems [34].…”

“…Thanks to that, the structure of the tree becomes simpler and robust to noise/outliers. In this step more than two nodes/leaves can be aggregated (lines [34][35][36]. If a particular leaf should be aggregated with its sibling (e.g., at level 3), it might happen that this sibling has its own children (at level 4).…”

Intelligent Energy Management of Electrical Power Systems

“…Regarding data, two papers are included in this Special Issue. The result of the analysis presented in [12] shows that revealing characteristics from smart meter data is feasible and the proposed machine learning methods have a good accuracy. Concerning the blockchain-based energy trading in the electrical power system, the survey [13] tackles challenges like energy transaction, consensus mechanism, and system optimization.…”

“…An intelligent energy management system implies also many other components. Some of them, like electric springs, smart meter, and data, are presented in [11][12][13]. Reference [11] introduces the state of the art of electric springs as a new solution for stabilizing power grid fed by intermittent renewable energy sources.…”

Weighted Random Forests to Improve Arrhythmia Classification

“…The second model has been built based on the framework described in [66], in order to construct a model of the support vector machine, the C-SVM function from the kernlab library was used. The linear, polynomial (degrees 1, 2, and 3) and radial (γ from 0.1 to 1, by 0.2) kernel functions were used and ε (which determines the margin width for which the error function is zero) was arbitrarily taken from the following set of {0.1, 0.3, 0.5, 0.7, 0.9}.…”