Performance comparison of neural networks for intelligent management of distributed generators in a distribution system

Nor Aira Zambri, Azah Mohamed, Mohd Z. Bin Che Wanik

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

The Multilayer Perceptron (MLP) neural network has been proven to be a very successful type of neural network in many applications. The MLP activation function is one of the important elements to be considered in neural network training in which proper selection of the activation function will give a huge impact on the network performance. This paper presents a comparative study of the four most commonly used activation functions in the neural network which include the sigmoid, hyperbolic tangent and linear functions used in the MLP neural network and the Gaussian function used in the Radial Basis Function (RBF) network for managing active and reactive power of distributed generation (DG) units in distribution systems. Simulation results show that the sigmoid activation functions give better performance in predicting the optimal power reference of the DG units. However, the RBF neural network gives the fastest conversion time compared to the MLP neural network.

Original languageEnglish
Pages (from-to)179-190
Number of pages12
JournalInternational Journal of Electrical Power and Energy Systems
Volume67
DOIs
Publication statusPublished - 2015
Externally publishedYes

Fingerprint

Neural networks
Multilayer neural networks
Chemical activation
Distributed power generation
Radial basis function networks
Network performance
Reactive power

Keywords

  • Activation function
  • Artificial neural network
  • Multilayer Perceptron
  • Radial basis function

ASJC Scopus subject areas

  • Energy Engineering and Power Technology
  • Electrical and Electronic Engineering

Cite this

@article{faadaa390b3c48a6884d6dc3a7880792,
title = "Performance comparison of neural networks for intelligent management of distributed generators in a distribution system",
abstract = "The Multilayer Perceptron (MLP) neural network has been proven to be a very successful type of neural network in many applications. The MLP activation function is one of the important elements to be considered in neural network training in which proper selection of the activation function will give a huge impact on the network performance. This paper presents a comparative study of the four most commonly used activation functions in the neural network which include the sigmoid, hyperbolic tangent and linear functions used in the MLP neural network and the Gaussian function used in the Radial Basis Function (RBF) network for managing active and reactive power of distributed generation (DG) units in distribution systems. Simulation results show that the sigmoid activation functions give better performance in predicting the optimal power reference of the DG units. However, the RBF neural network gives the fastest conversion time compared to the MLP neural network.",
keywords = "Activation function, Artificial neural network, Multilayer Perceptron, Radial basis function",
author = "Zambri, {Nor Aira} and Azah Mohamed and Wanik, {Mohd Z. Bin Che}",
year = "2015",
doi = "10.1016/j.ijepes.2014.11.005",
language = "English",
volume = "67",
pages = "179--190",
journal = "International Journal of Electrical Power and Energy Systems",
issn = "0142-0615",
publisher = "Elsevier Limited",

}

TY - JOUR

T1 - Performance comparison of neural networks for intelligent management of distributed generators in a distribution system

AU - Zambri, Nor Aira

AU - Mohamed, Azah

AU - Wanik, Mohd Z. Bin Che

PY - 2015

Y1 - 2015

N2 - The Multilayer Perceptron (MLP) neural network has been proven to be a very successful type of neural network in many applications. The MLP activation function is one of the important elements to be considered in neural network training in which proper selection of the activation function will give a huge impact on the network performance. This paper presents a comparative study of the four most commonly used activation functions in the neural network which include the sigmoid, hyperbolic tangent and linear functions used in the MLP neural network and the Gaussian function used in the Radial Basis Function (RBF) network for managing active and reactive power of distributed generation (DG) units in distribution systems. Simulation results show that the sigmoid activation functions give better performance in predicting the optimal power reference of the DG units. However, the RBF neural network gives the fastest conversion time compared to the MLP neural network.

AB - The Multilayer Perceptron (MLP) neural network has been proven to be a very successful type of neural network in many applications. The MLP activation function is one of the important elements to be considered in neural network training in which proper selection of the activation function will give a huge impact on the network performance. This paper presents a comparative study of the four most commonly used activation functions in the neural network which include the sigmoid, hyperbolic tangent and linear functions used in the MLP neural network and the Gaussian function used in the Radial Basis Function (RBF) network for managing active and reactive power of distributed generation (DG) units in distribution systems. Simulation results show that the sigmoid activation functions give better performance in predicting the optimal power reference of the DG units. However, the RBF neural network gives the fastest conversion time compared to the MLP neural network.

KW - Activation function

KW - Artificial neural network

KW - Multilayer Perceptron

KW - Radial basis function

UR - http://www.scopus.com/inward/record.url?scp=84916918379&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84916918379&partnerID=8YFLogxK

U2 - 10.1016/j.ijepes.2014.11.005

DO - 10.1016/j.ijepes.2014.11.005

M3 - Article

VL - 67

SP - 179

EP - 190

JO - International Journal of Electrical Power and Energy Systems

JF - International Journal of Electrical Power and Energy Systems

SN - 0142-0615

ER -