Impact of Data Quality on Day-ahead Photovoltaic Power Production Forecasting

Theocharides, Spyros; Tziolis, Georgios; Lopez‐Lorente, Javier; Makrides, George; Georghiou, George E.

doi:10.1109/pvsc43889.2021.9518471

Cited by 5 publications

(4 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For instance, Onal [1] created a PV system of dimensions

22

by interconnecting two panels in a series configuration and two panels in a parallel configuration, and then proceeded to simulate its performance by collecting solar irradiation data for an entire day. Along with this short-term trial, some experiments extend from one year to more than thirty years [32] , [33] . In some situations, researchers have collected data at different time intervals to enhance their understanding.…”

Section: Resultsmentioning

confidence: 99%

“…Authors varied time duration to yield a better outcome. Theocharides et al [32] conducted experiments using a one-year sample dataset and determined that ANN was the most suitable model for the given scenario. Duchaud et al [55] ran the experiment during the peak hours of the day.…”

Section: Resultsmentioning

confidence: 99%

“…Table 3 reveals that ANN wins in terms of best performing ML methods [38] , [32] , [50] , [54] , [58] , [61] . Jung et al [54] achieved 97%

score in their testing dataset using ANN, whereas Kim and Kim [36] achieved 96.42% and Zhao et al [50] achieved 82.90%

score.…”

Section: Resultsmentioning

confidence: 99%

See 2 more Smart Citations

A systematic review on predicting PV system parameters using machine learning

Jobayer

Shaikat

Rashid

et al. 2023

Heliyon

View full text Add to dashboard Cite

“…For instance, Onal [1] created a PV system of dimensions

22

Section: Resultsmentioning

confidence: 99%

Section: Resultsmentioning

confidence: 99%

See 1 more Smart Citation

A systematic review on predicting PV system parameters using machine learning

Jobayer

Shaikat

Rashid

et al. 2023

Heliyon

View full text Add to dashboard Cite

“…However, due to the intricate structure of these systems, providing an accurate forecast of the amount of electricity that will be generated by PV systems has proven to be difficult. The focus of current research has changed towards predicting methodologies that are more advanced and adaptable, making use of data-driven approaches that are based on machine learning algorithms [10][11][12][13][14].…”

Section: Introductionmentioning

confidence: 99%

PV generation forecasting utilizing a classification-only approach

Theocharides,

Makrides,

Georghiou

2024

EPJ Photovolt.

View full text Add to dashboard Cite

The increasing use of photovoltaic (PV) systems in electricity infrastructure poses new reliability challenges, as the supply of solar energy is primarily dependent on weather conditions. Consequently, to mitigate the issue, enhanced day-ahead PV production forecasts can be obtained by employing advanced machine learning techniques and reducing the uncertainty of solar irradiance predictions through statistical processing. The objective of this study was to present a methodology for accurately forecasting day-ahead PV production using novel machine learning techniques and a classification-only forecasting approach. Specifically, the central component of the proposed method is a classifier model based on an Extreme Gradient Boosting (XGBoost) ensemble algorithm that classifies the respective daily 30-min profiles of the forecasted global horizontal irradiance (GHI), the measured incident irradiance (Gi), and the AC power (PAC) into a predetermined number of classes. The formed classifier model was used as a dictionary to designate the newly arrived forecasted GHI to a particular class and ultimately identify the corresponding forecasted PAC. The results demonstrated that the proposed forecasting solution provided forecasts with a daily normalised root mean square error (nRMSE) of 8.20% and a mean absolute percentage error (MAPE) of 6.91% over the test set period of one year, while the model's reproducibility was also evaluated and confirmed. Additionally, a comprehensive evaluation based on clear-sky index categories revealed that the model's performance was notably accurate on clear-sky days, while maintaining acceptable accuracy levels on moderate and overcast days. These findings underscore the versatility and robustness of the proposed methodology in handling diverse weather conditions and hold promise for improved PV production forecasts.

show abstract