Home // International Journal On Advances in Systems and Measurements, volume 17, numbers 3 and 4, 2024 // View article

Day-ahead Forecasting Electricity Spot Prices in Norway with ARIMA, XGBoost, and LSTM Models

Authors:
Markus Jensen
Huamin Ren
Andrii Shalaginov

Keywords: Green Energy; Electricity Price Forecasting; Elspot prices; XGBoost; LSTM.

Abstract:
This paper comprehensively explores univariate and multivariate forecasting models for the Norwegian Elspot markets. As a leading renewable energy supplier with a high reliance on hydropower, Norway offers valuable insights into balancing renewable sources. The volatility of its electricity market, influenced by broader European trends, underscores the need for accurate forecasting. Day-ahead electricity price forecasts from the Elspot market are crucial for electricity producers and market operators, informing supply bids and dispatch schedules. This research includes experiments with advanced forecasting methods, combining machine learning and time series analysis to improve accuracy. We compare three models—ARIMA, XGBoost, and LSTM—across Norway's six Elspot markets. LSTM outperforms the other models in three specific zones, demonstrating its superior predictive performance. Future research will focus on enhancing model generalization.

Pages: 127 to 137

Copyright: Copyright (c) to authors, 2024. Used with permission.

Publication date: December 30, 2024

Published in: journal

ISSN: 1942-261x