Data driven models on load forecasting: Case study Algeria

Abstract

Databases are indispensable in many areas. This historical database contains a wealth of information, such as the energy. Energy is a vital resource in our daily lives, for both residential and industrial users. However, determining the exact amount of energy required to satisfy society's demands remains a difficult question. The only way for the moment to solve this problem is to create models using historical data. The Algerian provider has given this data in order to prepare for consumption according to the proposed models. This article focuses on hourly consumption data collected between 2008 and 2020, carefully collected to enable efficient energy modelling and enhance forecasting. This data serves as the basis for the development of statistical, mathematical and predictive models based on machine learning principles. To achieve this, an advanced analysis of the dataset is applied, using statistical techniques and concepts that will provide valuable insights and statistical knowledge, enabling them to make accurate predictions.

Keywords

• Energy
• Electricity
• Machine learning
• Statistical techniques
• Accurate predictions