Power demand forecasting using Machine Learning approaches

This thesis explores the application of various machine learning models to power demand forecasting, a crucial component in the efficient management of energy resources. The main goal of this thesis is to introduce the machine learning models used for forecasting power demand and compare those methods. Additional goals supporting the main goal are contributing to the optimization of the German energy grid, demonstration of the practical application using real data, and preparation for future research. Those goals were fulfilled by using literature and the author’s experience. Python programming language was used to implement the whole practical part. The outcome of this thesis is a concise comparison of different models, such as linear regression, XGBoost and long short-term memory neural network.