Electricity consumption forecast for Tarkwa using autoregressive integrated moving average and adaptive neuro fuzzy inference system
Electricity has become one of the inelastic goods in our world today. The proper functioning of most equipment today relies on electricity. Taking Tarkwa which is a mining community into consideration, the various mines, schools, shops, banks and other companies in the municipality massivel...
Main Authors: | , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Faculty of Technical Sciences in Cacak
2021-01-01
|
Series: | Serbian Journal of Electrical Engineering |
Subjects: | |
Online Access: | http://www.doiserbia.nb.rs/img/doi/1451-4869/2021/1451-48692101075O.pdf |
id |
doaj-f995e4ce859f47d9bcfc21979ea26db5 |
---|---|
record_format |
Article |
spelling |
doaj-f995e4ce859f47d9bcfc21979ea26db52021-04-09T09:47:38ZengFaculty of Technical Sciences in CacakSerbian Journal of Electrical Engineering1451-48692217-71832021-01-01181759410.2298/SJEE2101075O1451-48692101075OElectricity consumption forecast for Tarkwa using autoregressive integrated moving average and adaptive neuro fuzzy inference systemOfosu Robert Agyare0Odoi Benjamin1Asamoah Mercy2Faculty of Engineering, University of Mines and Technology, Tarkwa, GhanaFaculty of Engineering, University of Mines and Technology, Tarkwa, GhanaFaculty of Engineering, University of Mines and Technology, Tarkwa, GhanaElectricity has become one of the inelastic goods in our world today. The proper functioning of most equipment today relies on electricity. Taking Tarkwa which is a mining community into consideration, the various mines, schools, shops, banks and other companies in the municipality massively rely on electricity for their day to day running. Therefore, knowing the exact amount of electricity to produce and distribute for the smooth running of businesses and basic living is of great necessity. This study compared and formulated a model to forecast and predict the daily electrical energy consumption in Tarkwa for the year 2019. The data used was a monthly dataset for the year 2018 and it comprised of the temperature, wind speed, population and electricity consumption for Tarkwa. The methods used were Artificial Neuro-Fuzzy Inference System (ANFIS) and Autoregressive Integrated Moving Average (ARIMA). The ANFIS was used as a predictor to predict the electricity consumption based on the training and testing of the dependent and independent variables. The ARIMA was used to forecast the dependent and independent variables for 2019. These simulations were done using MATLand Minitab. The results of the analysis revealed that the training and testing dataset allowed ANFIS to learn and understand the system but the ANFIS could only forecast the 2019 electricity consumption after the input data to the system was changed to the ARIMA forecasted 2019 independent variables. It was observed that the amount of electricity consumed in 2019 increased by 14%.http://www.doiserbia.nb.rs/img/doi/1451-4869/2021/1451-48692101075O.pdfelectricity consumptionerror metricsarimaanfisforecast |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Ofosu Robert Agyare Odoi Benjamin Asamoah Mercy |
spellingShingle |
Ofosu Robert Agyare Odoi Benjamin Asamoah Mercy Electricity consumption forecast for Tarkwa using autoregressive integrated moving average and adaptive neuro fuzzy inference system Serbian Journal of Electrical Engineering electricity consumption error metrics arima anfis forecast |
author_facet |
Ofosu Robert Agyare Odoi Benjamin Asamoah Mercy |
author_sort |
Ofosu Robert Agyare |
title |
Electricity consumption forecast for Tarkwa using autoregressive integrated moving average and adaptive neuro fuzzy inference system |
title_short |
Electricity consumption forecast for Tarkwa using autoregressive integrated moving average and adaptive neuro fuzzy inference system |
title_full |
Electricity consumption forecast for Tarkwa using autoregressive integrated moving average and adaptive neuro fuzzy inference system |
title_fullStr |
Electricity consumption forecast for Tarkwa using autoregressive integrated moving average and adaptive neuro fuzzy inference system |
title_full_unstemmed |
Electricity consumption forecast for Tarkwa using autoregressive integrated moving average and adaptive neuro fuzzy inference system |
title_sort |
electricity consumption forecast for tarkwa using autoregressive integrated moving average and adaptive neuro fuzzy inference system |
publisher |
Faculty of Technical Sciences in Cacak |
series |
Serbian Journal of Electrical Engineering |
issn |
1451-4869 2217-7183 |
publishDate |
2021-01-01 |
description |
Electricity has become one of the inelastic goods in our world today. The
proper functioning of most equipment today relies on electricity. Taking
Tarkwa which is a mining community into consideration, the various mines,
schools, shops, banks and other companies in the municipality massively rely
on electricity for their day to day running. Therefore, knowing the exact
amount of electricity to produce and distribute for the smooth running of
businesses and basic living is of great necessity. This study compared and
formulated a model to forecast and predict the daily electrical energy
consumption in Tarkwa for the year 2019. The data used was a monthly dataset
for the year 2018 and it comprised of the temperature, wind speed,
population and electricity consumption for Tarkwa. The methods used were
Artificial Neuro-Fuzzy Inference System (ANFIS) and Autoregressive
Integrated Moving Average (ARIMA). The ANFIS was used as a predictor to
predict the electricity consumption based on the training and testing of the
dependent and independent variables. The ARIMA was used to forecast the
dependent and independent variables for 2019. These simulations were done
using MATLand Minitab. The results of the analysis revealed that the
training and testing dataset allowed ANFIS to learn and understand the
system but the ANFIS could only forecast the 2019 electricity consumption
after the input data to the system was changed to the ARIMA forecasted 2019
independent variables. It was observed that the amount of electricity
consumed in 2019 increased by 14%. |
topic |
electricity consumption error metrics arima anfis forecast |
url |
http://www.doiserbia.nb.rs/img/doi/1451-4869/2021/1451-48692101075O.pdf |
work_keys_str_mv |
AT ofosurobertagyare electricityconsumptionforecastfortarkwausingautoregressiveintegratedmovingaverageandadaptiveneurofuzzyinferencesystem AT odoibenjamin electricityconsumptionforecastfortarkwausingautoregressiveintegratedmovingaverageandadaptiveneurofuzzyinferencesystem AT asamoahmercy electricityconsumptionforecastfortarkwausingautoregressiveintegratedmovingaverageandadaptiveneurofuzzyinferencesystem |
_version_ |
1721532954696482816 |