A framework for classification of non-linear loads in smart grids using Artificial Neural Networks and Multi-Agent Systems

This paper proposes a general framework that uses the Artificial Neural Networks (ANNs) as a classification tool of nonlinear loads in a simulated smart grid environment by using Multi-Agent Systems (MAS). The increasing of communication and computation infrastructure on devices installed on modern...

Full description

Saved in:

Bibliographic Details
Published in:	Neurocomputing (Amsterdam) Vol. 170; pp. 328 - 338
Main Authors:	Saraiva, Filipe de O., Bernardes, Wellington M.S., Asada, Eduardo N.
Format:	Journal Article
Language:	English
Published:	Elsevier B.V 25.12.2015
Subjects:	Artificial Neural Networks Hybrid intelligent systems Multi-Agent Systems Non-linear loads classification Smart grids Smart grids Non-linear loads classification Artificial Neural Networks Hybrid intelligent systems Multi-Agent Systems
ISSN:	0925-2312, 1872-8286
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Description
Summary:	This paper proposes a general framework that uses the Artificial Neural Networks (ANNs) as a classification tool of nonlinear loads in a simulated smart grid environment by using Multi-Agent Systems (MAS). The increasing of communication and computation infrastructure on devices installed on modern power distribution systems allows new automated and coordinated control actions. This is mainly due to the ability to manage and process information and deploy actions in real-time mode. One important measurement tool is the smart meter, which will be present with all customers. Besides the measurement function, it has the communication feature and also some computational processing capability. Considering this base structure, the objective is to present methods to classify/identify nonlinear loads based only on current or voltage profiles measured by smart meters in this distributed computing environment. In this work, the MAS will manage the data and the tasks related to the classification and the ANN will perform the classification, both tools have been developed in JADE/JAVA and Matlab environment, respectively. Test case using 4000 input signals distributed in eight classes corresponding to nonlinear medical electromedical loads have been used and 98.7% of the samples have been identified correctly. •A framework for classification of non-linear loads in smart grids is proposed.•The framework uses multi-agent system to provide a communication infrastructure.•Artificial neural network was utilised as classification tool.•Two methods were presented and compared in terms of cost and sensitivity to faults.•Test case was defined using electrical loads collected from a hospital environment.
ISSN:	0925-2312 1872-8286
DOI:	10.1016/j.neucom.2015.02.090