Philip Dickemann

2024

This thesis explores the development and application of energy cost prediction (ECP) and electricity price forecasting (EPF) models amid recent socio-economic challenges as the Covid-19 pandemic or the Russian-Ukrainian conflict.
The research reviews scientific papers by systematic literature review and contrasts them to software solutions developed during the last decade. Our research reveals a significant lag in the adoption of leading-edge models, such as deep learning algorithms, which are prevalent in academia but scarcely represented in commercial software solutions. Only few solutions, as the one offered by QuantRisk, employ a custom array of computational intelligence models in their product, indicating a substantial disparity between academic research frontiers and software implementations.
This study concludes that closer collaboration and data exchange between academia and industry could enhance the accuracy of ECP and EPF models, bridging the gap between research and application.