Day-Ahead Forecast of Electric Vehicle Charging Demand with Deep Neural Networks
The increasing penetration rate of electric vehicles, associated with a growing charging demand, could induce a negative impact on the electric grid, such as higher peak power demand. To support the electric grid, and to anticipate those peaks, a growing interest exists for forecasting the day-ahead...
| Autores principales: | , , , , |
|---|---|
| Formato: | Artículo |
| Lenguaje: | English |
| Publicado: |
MDPI AG
2021-10-01
|
| Colección: | World Electric Vehicle Journal |
| Materias: | |
| Acceso en línea: | https://www.mdpi.com/2032-6653/12/4/178 |
| _version_ | 1827668930892660736 |
|---|---|
| author | Gilles Van Kriekinge Cedric De Cauwer Nikolaos Sapountzoglou Thierry Coosemans Maarten Messagie |
| author_facet | Gilles Van Kriekinge Cedric De Cauwer Nikolaos Sapountzoglou Thierry Coosemans Maarten Messagie |
| author_sort | Gilles Van Kriekinge |
| collection | DOAJ |
| description | The increasing penetration rate of electric vehicles, associated with a growing charging demand, could induce a negative impact on the electric grid, such as higher peak power demand. To support the electric grid, and to anticipate those peaks, a growing interest exists for forecasting the day-ahead charging demand of electric vehicles. This paper proposes the enhancement of a state-of-the-art deep neural network to forecast the day-ahead charging demand of electric vehicles with a time resolution of 15 min. In particular, new features have been added on the neural network in order to improve the forecasting. The forecaster is applied on an important use case of a local charging site of a hospital. The results show that the mean-absolute error (MAE) and root-mean-square error (RMSE) are respectively reduced by 28.8% and 19.22% thanks to the use of calendar and weather features. The main achievement of this research is the possibility to forecast a high stochastic aggregated EV charging demand on a day-ahead horizon with a MAE lower than 1 kW. |
| first_indexed | 2024-03-10T03:52:27Z |
| format | Article |
| id | doaj.art-ad51a7080ed24ff2bcac83964c5894b8 |
| institution | Directory Open Access Journal |
| issn | 2032-6653 |
| language | English |
| last_indexed | 2024-03-10T03:52:27Z |
| publishDate | 2021-10-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | World Electric Vehicle Journal |
| spelling | doaj.art-ad51a7080ed24ff2bcac83964c5894b82023-11-23T11:02:42ZengMDPI AGWorld Electric Vehicle Journal2032-66532021-10-0112417810.3390/wevj12040178Day-Ahead Forecast of Electric Vehicle Charging Demand with Deep Neural NetworksGilles Van Kriekinge0Cedric De Cauwer1Nikolaos Sapountzoglou2Thierry Coosemans3Maarten Messagie4EVERGi Research Group, MOBI Research Centre & ETEC Department, Vrije Universiteit Brussel (VUB), Pleinlaan 2, 1050 Brussel, BelgiumEVERGi Research Group, MOBI Research Centre & ETEC Department, Vrije Universiteit Brussel (VUB), Pleinlaan 2, 1050 Brussel, BelgiumEVERGi Research Group, MOBI Research Centre & ETEC Department, Vrije Universiteit Brussel (VUB), Pleinlaan 2, 1050 Brussel, BelgiumEVERGi Research Group, MOBI Research Centre & ETEC Department, Vrije Universiteit Brussel (VUB), Pleinlaan 2, 1050 Brussel, BelgiumEVERGi Research Group, MOBI Research Centre & ETEC Department, Vrije Universiteit Brussel (VUB), Pleinlaan 2, 1050 Brussel, BelgiumThe increasing penetration rate of electric vehicles, associated with a growing charging demand, could induce a negative impact on the electric grid, such as higher peak power demand. To support the electric grid, and to anticipate those peaks, a growing interest exists for forecasting the day-ahead charging demand of electric vehicles. This paper proposes the enhancement of a state-of-the-art deep neural network to forecast the day-ahead charging demand of electric vehicles with a time resolution of 15 min. In particular, new features have been added on the neural network in order to improve the forecasting. The forecaster is applied on an important use case of a local charging site of a hospital. The results show that the mean-absolute error (MAE) and root-mean-square error (RMSE) are respectively reduced by 28.8% and 19.22% thanks to the use of calendar and weather features. The main achievement of this research is the possibility to forecast a high stochastic aggregated EV charging demand on a day-ahead horizon with a MAE lower than 1 kW.https://www.mdpi.com/2032-6653/12/4/178aggregated charging demandday-ahead forecastelectric vehiclefeature importancerecurrent neural network |
| spellingShingle | Gilles Van Kriekinge Cedric De Cauwer Nikolaos Sapountzoglou Thierry Coosemans Maarten Messagie Day-Ahead Forecast of Electric Vehicle Charging Demand with Deep Neural Networks World Electric Vehicle Journal aggregated charging demand day-ahead forecast electric vehicle feature importance recurrent neural network |
| title | Day-Ahead Forecast of Electric Vehicle Charging Demand with Deep Neural Networks |
| title_full | Day-Ahead Forecast of Electric Vehicle Charging Demand with Deep Neural Networks |
| title_fullStr | Day-Ahead Forecast of Electric Vehicle Charging Demand with Deep Neural Networks |
| title_full_unstemmed | Day-Ahead Forecast of Electric Vehicle Charging Demand with Deep Neural Networks |
| title_short | Day-Ahead Forecast of Electric Vehicle Charging Demand with Deep Neural Networks |
| title_sort | day ahead forecast of electric vehicle charging demand with deep neural networks |
| topic | aggregated charging demand day-ahead forecast electric vehicle feature importance recurrent neural network |
| url | https://www.mdpi.com/2032-6653/12/4/178 |
| work_keys_str_mv | AT gillesvankriekinge dayaheadforecastofelectricvehiclechargingdemandwithdeepneuralnetworks AT cedricdecauwer dayaheadforecastofelectricvehiclechargingdemandwithdeepneuralnetworks AT nikolaossapountzoglou dayaheadforecastofelectricvehiclechargingdemandwithdeepneuralnetworks AT thierrycoosemans dayaheadforecastofelectricvehiclechargingdemandwithdeepneuralnetworks AT maartenmessagie dayaheadforecastofelectricvehiclechargingdemandwithdeepneuralnetworks |