Predicting Global Energy Consumption Through Data Mining Techniques

Abstract

With the explosion of the global population and technological progress, the electricity demand has skyrocketed. To ensure a consistent flow of power, it's essential to accurately predict energy usage ahead of time. Failure to do so could lead to potential outages and disrupt our daily lives. This research reviews previous research in the field of using data mining techniques to analyze electricity consumption data, optimize energy performance of buildings, and predict energy consumption in various industries. The study also aims to uncover patterns, correlations, and rules in electricity consumption worldwide using data mining techniques. The analysis is performed using various data mining techniques, such as simple K-Means and Expectation Maximization (EM). This selection is based on their prominent applications for similar problems in literature. The simple K-Means and EM algorithms showed successful outcomes on the dataset, which is evident in the clustering plots. Further, the performance of the Hierarchical Clustering algorithm was not up to the desired standard. This is probably due to the nature of the available dataset. These outcomes of the analysis will provide a valuable resource for decision-makers and stakeholders in the energy sector, as it will provide a deeper understanding of electricity consumption patterns and trends. This could lead to a sustainable future.