Customer active power consumption prediction for the next day based on historical profile

<p>Energy consumption prediction application is one of the most important fields<br />that is artificially controlled with Artificial Intelligence technologies to maintain<br />accuracy for electricity market costs reduction. This work presents a way to build<br />and apply a...

Full description

Saved in:
Bibliographic Details
Main Authors: Ahmad A. Goudah, Mohamed El-Habrouk, Dieter Schramm, Yasser G. Dessouky
Format: Article
Language:English
Published: Academy Publishing Center 2022-06-01
Series:Advances in Computing and Engineering
Subjects:
Online Access:http://apc.aast.edu/ojs/index.php/ACE/article/view/395
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849429936784277504
author Ahmad A. Goudah
Mohamed El-Habrouk
Dieter Schramm
Yasser G. Dessouky
author_facet Ahmad A. Goudah
Mohamed El-Habrouk
Dieter Schramm
Yasser G. Dessouky
author_sort Ahmad A. Goudah
collection DOAJ
description <p>Energy consumption prediction application is one of the most important fields<br />that is artificially controlled with Artificial Intelligence technologies to maintain<br />accuracy for electricity market costs reduction. This work presents a way to build<br />and apply a model to each costumer in residential buildings. This model is built by using Long Short Term Memory (LSTM) networks to address a demonstration of time-series prediction problem and Deep Learning to take into consideration the historical consumption of customers and hourly load profiles in order to predict future consumption. Using this model, the most probable sequence of a certain industrial customer’s consumption levels for a coming day is predicted. In the case of residential customers, determining the particular period of the prediction in terms of either a year or a month would be helpful and more accurate due to changes in consumption according to the changes in temperature and weather conditions in general. Both of them are used together in this research work to make a wide or narrow prediction window.</p><p><br />A test data set for a set of customers is used. Consumption readings for any<br />customer in the test data set applying LSTM model are varying between minimum and maximum values of active power consumption. These values are always alternating during the day according to customer consumption behavior. This consumption variation leads to leveling all readings to be determined in a finite set and deterministic values. These levels could be then used in building the prediction model. Levels of consumption’s are modeling states in the transition matrix. Twenty five readings are recorded per day on each hour and cover leap years extra ones. Emission matrix is built using twenty five values numbered from one to twenty five and represent the observations. Calculating probabilities of being in each level (node) is also covered. Logistic Regression Algorithm is used to determine the most probable nodes for the next 25 hours in case of residential or industrial customers.</p><p>Index Terms—Smart Grids, Load Forecasting, Consumption Prediction, Long Short Term Memory (LSTM), Logistic Regression Algorithm, Load Profile, Electrical Consumption.</p>
format Article
id doaj-art-38008e5bc52c405f9776f8b61fb07c61
institution Kabale University
issn 2735-5977
2735-5985
language English
publishDate 2022-06-01
publisher Academy Publishing Center
record_format Article
series Advances in Computing and Engineering
spelling doaj-art-38008e5bc52c405f9776f8b61fb07c612025-08-20T03:28:10ZengAcademy Publishing CenterAdvances in Computing and Engineering2735-59772735-59852022-06-0121174210.21622/ace.2022.02.1.017199Customer active power consumption prediction for the next day based on historical profileAhmad A. Goudah0Mohamed El-Habrouk1Dieter SchrammYasser G. DessoukyArab Academy for science, Technology and Maritime Transport, Information and Documentation CenterLecturer in Power Electronics at Alexandria University<p>Energy consumption prediction application is one of the most important fields<br />that is artificially controlled with Artificial Intelligence technologies to maintain<br />accuracy for electricity market costs reduction. This work presents a way to build<br />and apply a model to each costumer in residential buildings. This model is built by using Long Short Term Memory (LSTM) networks to address a demonstration of time-series prediction problem and Deep Learning to take into consideration the historical consumption of customers and hourly load profiles in order to predict future consumption. Using this model, the most probable sequence of a certain industrial customer’s consumption levels for a coming day is predicted. In the case of residential customers, determining the particular period of the prediction in terms of either a year or a month would be helpful and more accurate due to changes in consumption according to the changes in temperature and weather conditions in general. Both of them are used together in this research work to make a wide or narrow prediction window.</p><p><br />A test data set for a set of customers is used. Consumption readings for any<br />customer in the test data set applying LSTM model are varying between minimum and maximum values of active power consumption. These values are always alternating during the day according to customer consumption behavior. This consumption variation leads to leveling all readings to be determined in a finite set and deterministic values. These levels could be then used in building the prediction model. Levels of consumption’s are modeling states in the transition matrix. Twenty five readings are recorded per day on each hour and cover leap years extra ones. Emission matrix is built using twenty five values numbered from one to twenty five and represent the observations. Calculating probabilities of being in each level (node) is also covered. Logistic Regression Algorithm is used to determine the most probable nodes for the next 25 hours in case of residential or industrial customers.</p><p>Index Terms—Smart Grids, Load Forecasting, Consumption Prediction, Long Short Term Memory (LSTM), Logistic Regression Algorithm, Load Profile, Electrical Consumption.</p>http://apc.aast.edu/ojs/index.php/ACE/article/view/395smart gridsload forecastingconsumption predictionhidden markov modelviterbi algorithmloadprofileelectrical consumption
spellingShingle Ahmad A. Goudah
Mohamed El-Habrouk
Dieter Schramm
Yasser G. Dessouky
Customer active power consumption prediction for the next day based on historical profile
Advances in Computing and Engineering
smart grids
load forecasting
consumption prediction
hidden markov model
viterbi algorithm
loadprofile
electrical consumption
title Customer active power consumption prediction for the next day based on historical profile
title_full Customer active power consumption prediction for the next day based on historical profile
title_fullStr Customer active power consumption prediction for the next day based on historical profile
title_full_unstemmed Customer active power consumption prediction for the next day based on historical profile
title_short Customer active power consumption prediction for the next day based on historical profile
title_sort customer active power consumption prediction for the next day based on historical profile
topic smart grids
load forecasting
consumption prediction
hidden markov model
viterbi algorithm
loadprofile
electrical consumption
url http://apc.aast.edu/ojs/index.php/ACE/article/view/395
work_keys_str_mv AT ahmadagoudah customeractivepowerconsumptionpredictionforthenextdaybasedonhistoricalprofile
AT mohamedelhabrouk customeractivepowerconsumptionpredictionforthenextdaybasedonhistoricalprofile
AT dieterschramm customeractivepowerconsumptionpredictionforthenextdaybasedonhistoricalprofile
AT yassergdessouky customeractivepowerconsumptionpredictionforthenextdaybasedonhistoricalprofile