Abstract
Deep learning models have the potential to advance the short-term decision-making of electricity market participants and system operators by capturing the complex dependences and uncertainties of power system operation. Currently, however, the adoption of global deep learning models for multivariate energy forecasting in power systems is far behind the developments in the deep learning research field. In this context, the objectives of this study are to review recent developments in the field of probabilistic, multivariate, and multihorizon time series forecasting and empirically evaluate the performance of novel global deep learning models for forecasting wind and solar generation, electricity load, and wholesale electricity price for intraday and day-ahead time horizons. Two forecast types, deterministic and probabilistic forecasts, are studied. The evaluation data consist of real-world datasets with hourly resolution at the levels of an individual customer and regional and national electricity market bidding zones. The model evaluation criteria include achievable levels of forecasting accuracy and uncertainty risks, hyperparameter sensitivity, the effect of exogenous variables and fieldwise dataset split, and run-time efficiency factors, such as memory utilization, simulation time, electricity consumption, and convergence rate. We conclude that the performance of the global models is more beneficial for intraday forecasts of heterogeneous datasets with nonuniform patterns of time series, but can be affected by the hyperparameter sensitivity and hardware limitations with the growth of dataset dimensionality. The results can serve as a reference point for the quantitative evaluation of deep learning models for probabilistic multivariate energy forecasting in power systems.
Highlights
The short-term forecasting of energy time series, such as wind and solar energy, electricity load and price, is at the core of electric power system trading and operation
The long- and short-term time series network (LSTNet) and dual selfattention network (DSANet) models belong to the class of many-to-one models that predict one value that is at the horizon distance from an input sequence, whereas DeepAR and deep temporal convolutional network (DeepTCN) are many-to-many models predicting a sequence of a horizon length ahead given the input sequence
This study bridges the gap between the adoption of novel global deep-learning-based models for probabilistic multivariate forecasting in the deep learning community and the applicability of these methods for energy forecasting
Summary
The short-term forecasting of energy time series, such as wind and solar energy, electricity load and price, is at the core of electric power system trading and operation It provides the electricity market participants and system operators with information on the hours and days to enable cost-efficient market bidding and operating reserve procurement and detecting network congestion. The operational predictability of modern power systems is being challenged by intermittency, uncertainty, and stochasticity as the installed capacity of renewable energy sources (RES) increases and new distributed energy resources are introduced into the existing power networks To adapt to these decarbonization and decentralization trends and support the risk-aware decision-making of power system actors, the present energy forecasting approaches should be improved to estimate the prediction uncertainties and leverage large amounts of data with complex multivariate dependences [1]. These dependences are limited to linear relationships [42]
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.