University of Twente Student Theses


Dependable Probabilistic Energy Forecasting of Solar Energy for Energy Management Systems

Doornkamp, C.J. (2023) Dependable Probabilistic Energy Forecasting of Solar Energy for Energy Management Systems.

[img] PDF
Abstract:In this thesis, a methodology has been researched and developed to aid in the decision-making process of Energy Management Systems (EMS). Current solar energy forecasting methods forecast the expected energy production of solar panels at certain points in the future, called point forecasts. However, due to the unpredictable nature of the weather, point forecasts cannot fully describe the future. Therefore, a different modelling method has been researched and developed that takes this uncertainty into account and provides additional information to the EMS. This is done by forecasting a probability distribution of the expected solar energy generation of the PV installation. Given the different nature of probabilistic forecasts, the traditional methodology and data used for point forecasts need not apply to probabilistic forecasts. Therefore, new forecasting methods and data sources have been investigated that are best suited for probabilistic solar power forecasting, without it being based on assumptions on point forecasts. Additionally, different use cases and methods are provided in this thesis on how these probabilistic forecasts can be interpreted to enhance an EMS’ operation and how it allows for unique applications that are not possible using point forecasts.
Item Type:Essay (Master)
Faculty:EEMCS: Electrical Engineering, Mathematics and Computer Science
Subject:31 mathematics, 50 technical science in general
Programme:Embedded Systems MSc (60331)
Link to this item:
Export this item as:BibTeX
HTML Citation
Reference Manager


Repository Staff Only: item control page