Midterm Load Forecasting Analysis For Erbil Governorate Based On Predictive Models

Abstract

   Electrical power supply is becoming more and more complex as a result of expansion, growing population, and unsuitable planning of administration and peoples. Electrical power load forecasting may be defined as the process of predicting electrical load values for future of the system with respect to current demands. This analysis is an important procedure for the power system planners and the demand controllers to ensure that the system can generate sufficient of electricity for different kinds of terms such as short, medium and long term load forecasting. The forecasting analysis allows us to manage the electrical loads with the increasing demand. For that purpose, we have used some predictive models to analyze of electrical load forecasting for Erbil Governorate in Iraq. This analysis helps us to manage our planning better, arrange system maintenance plan and enhance fuel control. This study raises an attempt for forecasting the peak (upper limit) monthly demand of electric power for one year ahead.  Simple linear regression model and Auto Regressive Integrated Moving Average model were applied as forecasting models for a power consumer’s dataset for the purpose of predicting forthcoming year   electricity load demand.  ,  also Forecasting models are then validated using some indicators, indicator used is Root Mean Square Error (RMSE) ,which is conceded a statistic metric that is commonly used for  accuracy evaluation of LF methods,   and Mean Absolute Error (MAE) both used as a forecasting accuracy criteria.