Short-Term Net Load Forecasting with Singular Spectrum Analysis and LSTM Neural Networks
Short-term electricity load forecasting is key to the safe, reliable, and economical operation of power systems. An important challenge that arises with high-frequency load series, e.g., hourly load, is how to deal with the complex seasonal patterns that are present. Standard approaches suggest eith...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-07-01
|
| Series: | Energies |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1996-1073/14/14/4107 |
| _version_ | 1827687664540712960 |
|---|---|
| author | Akylas Stratigakos Athanasios Bachoumis Vasiliki Vita Elias Zafiropoulos |
| author_facet | Akylas Stratigakos Athanasios Bachoumis Vasiliki Vita Elias Zafiropoulos |
| author_sort | Akylas Stratigakos |
| collection | DOAJ |
| description | Short-term electricity load forecasting is key to the safe, reliable, and economical operation of power systems. An important challenge that arises with high-frequency load series, e.g., hourly load, is how to deal with the complex seasonal patterns that are present. Standard approaches suggest either removing seasonality prior to modeling or applying time series decomposition. This work proposes a hybrid approach that combines Singular Spectrum Analysis (SSA)-based decomposition and Artificial Neural Networks (ANNs) for day-ahead hourly load forecasting. First, the trajectory matrix of the time series is constructed and decomposed into trend, oscillating, and noise components. Next, the extracted components are employed as exogenous regressors in a global forecasting model, comprising either a Multilayer Perceptron (MLP) or a Long Short-Term Memory (LSTM) predictive layer. The model is further extended to include exogenous features, e.g., weather forecasts, transformed via parallel dense layers. The predictive performance is evaluated on two real-world datasets, controlling for the effect of exogenous features on predictive accuracy. The results showcase that the decomposition step improves the relative performance for ANN models, with the combination of LSTM and SAA providing the best overall performance. |
| first_indexed | 2024-03-10T09:41:14Z |
| format | Article |
| id | doaj.art-88028652d3ce4dd391f64bf4c4b89f34 |
| institution | Directory Open Access Journal |
| issn | 1996-1073 |
| language | English |
| last_indexed | 2024-03-10T09:41:14Z |
| publishDate | 2021-07-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Energies |
| spelling | doaj.art-88028652d3ce4dd391f64bf4c4b89f342023-11-22T03:40:11ZengMDPI AGEnergies1996-10732021-07-011414410710.3390/en14144107Short-Term Net Load Forecasting with Singular Spectrum Analysis and LSTM Neural NetworksAkylas Stratigakos0Athanasios Bachoumis1Vasiliki Vita2Elias Zafiropoulos3Center PERSEE, MINES ParisTech, PSL University, 1 Rue Claude Daunesse, 06904 Sophia Antipolis, FranceUbitech Energy, Koningin Astridlaan 59b, 1780 Brussels, BelgiumDepartment of Electrical and Electronic Engineering Educators, A.S.PE.T.E.—School of Pedagogical and Technological Education, N. Heraklion, 14121 Athens, GreeceUbitech Energy, Koningin Astridlaan 59b, 1780 Brussels, BelgiumShort-term electricity load forecasting is key to the safe, reliable, and economical operation of power systems. An important challenge that arises with high-frequency load series, e.g., hourly load, is how to deal with the complex seasonal patterns that are present. Standard approaches suggest either removing seasonality prior to modeling or applying time series decomposition. This work proposes a hybrid approach that combines Singular Spectrum Analysis (SSA)-based decomposition and Artificial Neural Networks (ANNs) for day-ahead hourly load forecasting. First, the trajectory matrix of the time series is constructed and decomposed into trend, oscillating, and noise components. Next, the extracted components are employed as exogenous regressors in a global forecasting model, comprising either a Multilayer Perceptron (MLP) or a Long Short-Term Memory (LSTM) predictive layer. The model is further extended to include exogenous features, e.g., weather forecasts, transformed via parallel dense layers. The predictive performance is evaluated on two real-world datasets, controlling for the effect of exogenous features on predictive accuracy. The results showcase that the decomposition step improves the relative performance for ANN models, with the combination of LSTM and SAA providing the best overall performance.https://www.mdpi.com/1996-1073/14/14/4107LSTMshort-term load forecastingsingular spectrum analysistime series decomposition |
| spellingShingle | Akylas Stratigakos Athanasios Bachoumis Vasiliki Vita Elias Zafiropoulos Short-Term Net Load Forecasting with Singular Spectrum Analysis and LSTM Neural Networks Energies LSTM short-term load forecasting singular spectrum analysis time series decomposition |
| title | Short-Term Net Load Forecasting with Singular Spectrum Analysis and LSTM Neural Networks |
| title_full | Short-Term Net Load Forecasting with Singular Spectrum Analysis and LSTM Neural Networks |
| title_fullStr | Short-Term Net Load Forecasting with Singular Spectrum Analysis and LSTM Neural Networks |
| title_full_unstemmed | Short-Term Net Load Forecasting with Singular Spectrum Analysis and LSTM Neural Networks |
| title_short | Short-Term Net Load Forecasting with Singular Spectrum Analysis and LSTM Neural Networks |
| title_sort | short term net load forecasting with singular spectrum analysis and lstm neural networks |
| topic | LSTM short-term load forecasting singular spectrum analysis time series decomposition |
| url | https://www.mdpi.com/1996-1073/14/14/4107 |
| work_keys_str_mv | AT akylasstratigakos shorttermnetloadforecastingwithsingularspectrumanalysisandlstmneuralnetworks AT athanasiosbachoumis shorttermnetloadforecastingwithsingularspectrumanalysisandlstmneuralnetworks AT vasilikivita shorttermnetloadforecastingwithsingularspectrumanalysisandlstmneuralnetworks AT eliaszafiropoulos shorttermnetloadforecastingwithsingularspectrumanalysisandlstmneuralnetworks |