Half-hourly electricity price prediction model with explainable-decomposition hybrid deep learning approach
Ghimire, Sujan; Deo, Ravinesh C.; Hopf, Konstantin; u. a. (2025): Half-hourly electricity price prediction model with explainable-decomposition hybrid deep learning approach, in: Energy and AI, Amsterdam: Elsevier BV, Jg. 20, Nr. 100492, S. 1–34, doi: 10.1016/j.egyai.2025.100492.
Details
- Fakultät/Lehrstuhl
- Autorin(nen)/Autor(en)
- Ghimire, Sujan; Deo, Ravinesh C.; Hopf, Konstantin; Liu, Hangyue; Casillas-Pérez, David; Helwig, Andreas; Prasad, Salvin S.; Pérez-Aracil, J.; Barua, Prabal Datta; Salcedo-Sanz, Sancho
- Titel der Zeitschrift
- Energy and AI
- ISSN
- 2666-5468
- Verlagsangabe
- Elsevier BV : Amsterdam
- Erscheinungsjahr
- 2025
- Jahrgang
- 20
- Heft
- 100492
- Sprache(n)
- Englisch
- DOI
- Abstract
Accurate prediction of electricity price (𝐸𝑃 ) is crucial for energy utilities and grid operators for enhancing the energy trading, grid stability studies, resource allocations and pricing strategies, thereby improving the overall grid reliability, efficiency, and cost-effectiveness. This study introduces a novel D3Net model for halfhourly 𝐸𝑃 prediction, integrating Seasonal-Trend decomposition using LOESS (STL) and Variational Mode Decomposition (VMD) with Multi-Layer Perceptron (MLP), Random Forest Regression (RFR), and Tabular Neural Network (TabNet). The methodology involves applying STL to the 𝐸𝑃 time-series to extract trend, seasonal, and residual components. The trend is predicted using an MLP model, the seasonal component is further decomposed with VMD into 20 Variational Mode Functions (VMFs) and predicted using an RFR model, and the residual component is decomposed with VMD and predicted using the TabNet model. Input features are identified using the Partial Autocorrelation Function, and models are optimized using the Optuna algorithm.
- Freie Schlagwort(e)
- Tabular neural networkSHAPOptuna algorithmDeep learningMachine learningConvolutional neural networkLIME
- Peer Reviewed
- ja
- Internationale Verbreitung
- ja
- Open-Access-Zeitschrift
- ja
- Dokumenttyp
- Artikel in einer Zeitschrift / Postprint
- URI
- Datum der Freischaltung
- 01.04.2025