Analysing the Influence of Factors on Short-Term Electricity Consumption Forecasting: Insights from Historical Data

Abstract

This research investigates the forecast of short-term electricity consumption. The impact of various factors on prediction accuracy is investigated using past data. It is observed that certain factors exhibit higher importance in forecasting accuracy, with their significance diminishing over time. Two models were employed for prediction: one utilizing 24-hour past data and the other 72-hour past data for forecasting the subsequent 24-hour demand. It was unexpected to discover that the model that only used historical demand data performed better than predicted. This research contributes to a better understanding of short-term electricity demand forecasting, which is crucial for efficient energy planning.