Long-term Energy Demand Analysis using Machine Learning Algorithms: A Case Study in Bangladesh

In each energy sector from production to consumption machine learning and statistical models have intimidated the power system for the past few decades. Among these, energy demand analysis is one of the approved research projects in Bangladesh due to its necessity. Such a study introduces energy consumption forecasting along with necessary environmental featured variables. These energy demands vary depending on short-term, medium-term, and long-term analysis. Although, this paper utilizes long-term energy demand research while predicting energy consumption. The dataset in the study includes three years of monthly data consisting of more than one million samples and diverse consumer types with versatile features of the capital of Bangladesh, Dhaka (mainly Uttara). Different machine learning models like; Random Forest, k-neighrest Neighbors regression, Extreme Gradient Boosting Method, and Light Gradient Boosting Method have been used and in order to evaluate them various performance constraints have been utilized. Among the models, KNN has performed considerably well. It has the result of about 0.9447 as R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup>, 163.9 as RMSE, and 28.7 as MAE. Such study will surely subsidize upgrading to the power system management of Bangladesh.