Page 136 - IJEEE-2022-Vol18-ISSUE-1

Received: 05 February 2022 Revised: 15 April 2022 Accepted: 20 April 2022
DOI: 10.37917/ijeee.18.1.14
Vol. 18| Issue 1| June 2022
? Open Access

Iraqi Journal for Electrical and Electronic Engineering

Original Article

Five-Component Load Forecast in Residential Sector
Using Smart Methods

Yamama A. I. Al-Nasiri*1, Hussein Al-bayaty1, Majid S.M. Al-Hafidh2
1 Department of Electrical Engineering, University of Kirkuk, Iraq
2 Department of Electrical Engineering, Mosul University, Iraq

Correspondence
* Yamama A. Al-Nasiri
Department of Electrical Engineering,
University of Kirkuk, Iraq
Email: yamama.elc@uokirkuk.edu.iq

Abstract
The electrical load is affected by the weather conditions in many countries as well as in Iraq. The weather-sensitive electrical
load is, usually, divided into two components, a weather-sensitive component, and a weather-insensitive component. The
research provides a method for separating the weather-sensitive electrical load into five components. and aims to prove the
efficiency of the five-component load Forecasting model. The artificial neural network was used to predict the weather-sensitive
electrical load using the MATLAB R17a software. Weather data and loads were used for one year for Mosul City. The
performance of the artificial neural network was evaluated using the mean squared error and the mean absolute percentage
error. The results indicate the accuracy of the prediction model used, MAPE equal to 0.0402.
KEYWORDS: Artificial Neural Network, Mean Squared Error, Weather Sensitive Load, Medium Term Load Forecast.

I. INTRODUCTION machine [20][16], fuzzy logic [8][11], and genetic algorithm
[7][21]. Electrical loads in many countries, including Iraq,
Accurate forecasting of electrical loads is essential to the are affected by weather conditions, especially temperatures.
electrical system for various purposes, including load Weather-sensitive loads are studied by dividing them into
management, plant expansion planning, intelligent two components: weather-sensitive and other weather-
operation, and accurate electrical energy pricing [1][2]. The insensitive. Countries are increasingly suffering from the
importance of forecasting increases with the increasing use impact of electrical loads due to weather conditions in winter
of renewable energy [3][4]. Load prediction (forecast) is to and summer. Since electrical loads differ in winter and
obtain future information based on previous readings that summer, it is best to separate weather-sensitive loads into
helps in taking appropriate action to achieve a balance three components. The first is not affected by weather
between generation and consumption [5][6][7]. In addition conditions (base component). The second is affected by high
to avoiding the disruption of loads at low prediction, and temperatures. The third is affected by low temperatures.
wastage problems in obstetrics at high prediction. Modern Separating electrical loads into three components leads to
technologies such as demand-side management and smart load management with greater accuracy and efficiency
grid have made accurate electrical load prediction more [22][23][24]. The research aims to prove the efficiency of the
important [8][9]. On the other hand, due to the increasing use five-component load prediction model and the main points
of renewable energy sources, accurate forecasting of for its application in predicting the loads as a general case.
electrical loads ensures optimum energy savings, battery
operation, energy management, and storage [10][11]. II. RESEARCH METHOD
Electrical load forecasting is classified according to the
forecast horizon into three types: short-term forecasting (a In current forecasting methods that predict weather-
few hours to one week), medium-term forecasting (weeks to sensitive loads, the electrical load divide into two
a year), and long-term forecasting (more than a year) components. A weather-insensitive component and a
[12][10]. Prediction of medium-term load forecast has not weather-sensitive component, this component includes the
been studied extensively, compared to the prediction of summer and winter electrical load. Electrical loads vary in
short-term or long-term loads [15]. Many traditional and summer and winter as well as periods. Therefore, separating
smart methods have been used to predict the electrical load the weather-sensitive component gives better accuracy,
such as time series analysis [16][17], artificial neural when it is divided into two components, one for summer and
networks [13][18], wavelet transform [19], vector support the other for winter.

This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and
reproduction in any medium, provided the original work is properly cited.
© 2022 The Authors. Iraqi Journal for Electrical and Electronic Engineering by College of Engineering, University of Basrah.

https://doi.org/10.37917/ijeee.18.1.14 https://www.ijeee.edu.iq 132