Day-Ahead Photovoltaic Forecasting: A Comparison of the Most Effective Techniques
We compare the 24-hour ahead forecasting performance of two methods commonly used for the prediction of the power output of photovoltaic systems. Both methods are based on Artificial Neural Networks (ANN), which have been trained on the same dataset, thus enabling a much-needed homogeneous compariso...
Main Authors: | , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2019-04-01
|
Series: | Energies |
Subjects: | |
Online Access: | https://www.mdpi.com/1996-1073/12/9/1621 |
_version_ | 1811301164689391616 |
---|---|
author | Alfredo Nespoli Emanuele Ogliari Sonia Leva Alessandro Massi Pavan Adel Mellit Vanni Lughi Alberto Dolara |
author_facet | Alfredo Nespoli Emanuele Ogliari Sonia Leva Alessandro Massi Pavan Adel Mellit Vanni Lughi Alberto Dolara |
author_sort | Alfredo Nespoli |
collection | DOAJ |
description | We compare the 24-hour ahead forecasting performance of two methods commonly used for the prediction of the power output of photovoltaic systems. Both methods are based on Artificial Neural Networks (ANN), which have been trained on the same dataset, thus enabling a much-needed homogeneous comparison currently lacking in the available literature. The dataset consists of an hourly series of simultaneous climatic and PV system parameters covering an entire year, and has been clustered to distinguish sunny from cloudy days and separately train the ANN. One forecasting method feeds only on the available dataset, while the other is a hybrid method as it relies upon the daily weather forecast. For sunny days, the first method shows a very good and stable prediction performance, with an almost constant Normalized Mean Absolute Error, NMAE%, in all cases (1% < NMAE% < 2%); the hybrid method shows an even better performance (NMAE% < 1%) for two of the days considered in this analysis, but overall a less stable performance (NMAE% > 2% and up to 5.3% for all the other cases). For cloudy days, the forecasting performance of both methods typically drops; the performance is rather stable for the method that does not use weather forecasts, while for the hybrid method it varies significantly for the days considered in the analysis. |
first_indexed | 2024-04-13T07:04:02Z |
format | Article |
id | doaj.art-13ce2f01855b4d9f89895286eea127b5 |
institution | Directory Open Access Journal |
issn | 1996-1073 |
language | English |
last_indexed | 2024-04-13T07:04:02Z |
publishDate | 2019-04-01 |
publisher | MDPI AG |
record_format | Article |
series | Energies |
spelling | doaj.art-13ce2f01855b4d9f89895286eea127b52022-12-22T02:57:02ZengMDPI AGEnergies1996-10732019-04-01129162110.3390/en12091621en12091621Day-Ahead Photovoltaic Forecasting: A Comparison of the Most Effective TechniquesAlfredo Nespoli0Emanuele Ogliari1Sonia Leva2Alessandro Massi Pavan3Adel Mellit4Vanni Lughi5Alberto Dolara6Department of Energy, Politecnico di Milano, 20156 Milano, ItalyDepartment of Energy, Politecnico di Milano, 20156 Milano, ItalyDepartment of Energy, Politecnico di Milano, 20156 Milano, ItalyDepartment of Engineering and Architecture, Università degli Studi di Trieste, 34127 Trieste, ItalyRenewable Energy Laboratory, Jijel University, Jijel 18000, AlgeriaDepartment of Engineering and Architecture, Università degli Studi di Trieste, 34127 Trieste, ItalyDepartment of Energy, Politecnico di Milano, 20156 Milano, ItalyWe compare the 24-hour ahead forecasting performance of two methods commonly used for the prediction of the power output of photovoltaic systems. Both methods are based on Artificial Neural Networks (ANN), which have been trained on the same dataset, thus enabling a much-needed homogeneous comparison currently lacking in the available literature. The dataset consists of an hourly series of simultaneous climatic and PV system parameters covering an entire year, and has been clustered to distinguish sunny from cloudy days and separately train the ANN. One forecasting method feeds only on the available dataset, while the other is a hybrid method as it relies upon the daily weather forecast. For sunny days, the first method shows a very good and stable prediction performance, with an almost constant Normalized Mean Absolute Error, NMAE%, in all cases (1% < NMAE% < 2%); the hybrid method shows an even better performance (NMAE% < 1%) for two of the days considered in this analysis, but overall a less stable performance (NMAE% > 2% and up to 5.3% for all the other cases). For cloudy days, the forecasting performance of both methods typically drops; the performance is rather stable for the method that does not use weather forecasts, while for the hybrid method it varies significantly for the days considered in the analysis.https://www.mdpi.com/1996-1073/12/9/1621neural networksday-ahead forecastingPV systemmicro-grid |
spellingShingle | Alfredo Nespoli Emanuele Ogliari Sonia Leva Alessandro Massi Pavan Adel Mellit Vanni Lughi Alberto Dolara Day-Ahead Photovoltaic Forecasting: A Comparison of the Most Effective Techniques Energies neural networks day-ahead forecasting PV system micro-grid |
title | Day-Ahead Photovoltaic Forecasting: A Comparison of the Most Effective Techniques |
title_full | Day-Ahead Photovoltaic Forecasting: A Comparison of the Most Effective Techniques |
title_fullStr | Day-Ahead Photovoltaic Forecasting: A Comparison of the Most Effective Techniques |
title_full_unstemmed | Day-Ahead Photovoltaic Forecasting: A Comparison of the Most Effective Techniques |
title_short | Day-Ahead Photovoltaic Forecasting: A Comparison of the Most Effective Techniques |
title_sort | day ahead photovoltaic forecasting a comparison of the most effective techniques |
topic | neural networks day-ahead forecasting PV system micro-grid |
url | https://www.mdpi.com/1996-1073/12/9/1621 |
work_keys_str_mv | AT alfredonespoli dayaheadphotovoltaicforecastingacomparisonofthemosteffectivetechniques AT emanueleogliari dayaheadphotovoltaicforecastingacomparisonofthemosteffectivetechniques AT sonialeva dayaheadphotovoltaicforecastingacomparisonofthemosteffectivetechniques AT alessandromassipavan dayaheadphotovoltaicforecastingacomparisonofthemosteffectivetechniques AT adelmellit dayaheadphotovoltaicforecastingacomparisonofthemosteffectivetechniques AT vannilughi dayaheadphotovoltaicforecastingacomparisonofthemosteffectivetechniques AT albertodolara dayaheadphotovoltaicforecastingacomparisonofthemosteffectivetechniques |