Conference paper for 2024 IEEE International Conference on Communications, Control, and Computing Technologies for Smart Grids (SmartGridComm) – Workshop on Data-driven Methods for Distribution Grid Monitoring, Operation and Planning. DOI: 10.1109/SmartGridComm60555.2024.10738097
Abstract:
Smart grid developments have gained significant attention due to their potential to optimise energy consumption and reduce environmental impacts. For this reason, it is crucial to forecast future state conditions such as power, temperatures, heat, or SOC (state-of-charge) to make the most accurate and suitable control decision depending on the context and need. Since many processes are hard to model, the forecasting task can be executed by exploiting the advantages of machine learning models such as LSTM, transformer, Autoformer, or CNN. Comparing the results to previous works, we can state that our best model also outperforms state-of-the-art and state-research forecasting methods for continuous variables like temperatures. The model’s ability to accurately predict future states allows for more informed and adaptive control decisions, leading to enhanced energy efficiency, reduced environmental impact, and improved grid stability.
