Monthly Peak-load Demand Forecasting for Sulaimani Governorate Using Different Weather Conditions Based on ANN Model

Abstract

Mid-term forecasting of load demand is necessary for the correct operation of electric utilities. There is an on-going attention toward putting new approaches to the task. Recently, Artificial Neural Network has played a successful role in various applica-tions. This paper is presents a monthly peak-load demand forecasting for Sulaimani (located in northern Iraq) using the most widely used traditional method based on an artificial natural network, the performance of the model is tested on the actual historical monthly demand of the Governorate for the years 2013 to 2017. The standard mean absolute percentage error method is used to evaluate the accuracy of forecasting models, the results obtained shows a very good estimation of the load. The mean absolute percentage error MAPE is 0.056.

Keywords

• midterm monthly load forecast
• artificial neural network
• multi-layer perceptron (mlp)
• actual load
• predicted load. yearly ahead.