A deep learning model for intra-day forecasting of solar irradiance using satellite-based estimations in the vicinity of a PV power plant

Pérez, Emilio; Pérez, J. M. Alonso; Segarra-Tamarit, Jorge; Beltrán, Héctor

doi:10.1016/j.solener.2021.02.033

Cited by 43 publications

(14 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Hence, the incorporated NWP forecasts are those from the European Centre for Medium‐Range Weather Forecasts (ECMWF), [ 52 ] which are obtained every 12 h and have a 10 day forecasting horizon with 1 h time steps for the first 90 h. Forecast of the surface net solar radiation are obtained from the high‐resolution forecast: Atmospheric Model high‐resolution 10‐day forecast (HRES), which has a spatial resolution of 0.1° for both latitude and longitude. Moreover, the DNN proposed in Pérez et al [ 53 ] is used to obtain the intraday forecasts. This DNN uses as its main input the results from the Surface Insolation under Clear and Cloudy Skies algorithm, [ 54 ] which provides estimated irradiance images from satellite data around the target location.…”

Section: Control Methodology To Firm the Pv Plant Production With Bat...mentioning

confidence: 99%

Influence of the Intraday Electricity Market Structure on the Degradation of Li‐Ion Batteries Used to Firm Photovoltaic Production

et al. 2022

Self Cite

View full text Add to dashboard Cite

This article considers the introduction of Li‐ion batteries in photovoltaic power plants to firm their energy production and analyzes the dependence of their degradation on the structure of the electricity market where the power production is traded. The operation of the batteries is decided as a result of successive optimization problems that benefit from the use of deep‐learning‐based irradiance forecasting tools with low prediction error, which allows the batteries to keep a small size. In addition, state‐of‐the‐art battery aging models are handled to derive a realistic lifetime prognosis. The simulation results obtained by using real data from three European locations, which have different irradiance patterns and market structures, show how the proposed control strategy makes it possible to decouple the saturation rate of the batteries from the climatic conditions of the plant location. Furthermore, regarding the market structure, the results show that the shorter the energy block and the closer the lead time, the lower is the degradation of the batteries.

show abstract

Section: Control Methodology To Firm the Pv Plant Production With Bat...mentioning

confidence: 99%

Influence of the Intraday Electricity Market Structure on the Degradation of Li‐Ion Batteries Used to Firm Photovoltaic Production

et al. 2022

Self Cite

View full text Add to dashboard Cite

show abstract

“…More recently, convolutional neural networks have been used to predict future frames or future irradiance levels from a sequence of past satellite images (Pérez et al [45], Si et al [48], Nielsen et al [33]). Contrary to sky images which correlate well with their corresponding irradiance level, raw satellite observations are often supplemented by additional information on the current solar irradiance value to improve the performance.…”

Section: Satellite Imagerymentioning

confidence: 99%

“…Contrary to sky images which correlate well with their corresponding irradiance level, raw satellite observations are often supplemented by additional information on the current solar irradiance value to improve the performance. Pérez et al [45] do not rely on ground measurements but use physics-based surface solar irradiance maps as a 2D input to a CNN model. However, they also evaluate the potential benefit of integrating past irradiance measurements and corresponding clear-sky irradiance levels to calibrate the model.…”

Section: Satellite Imagerymentioning

confidence: 99%

Omnivision forecasting: combining satellite observations with sky images for improved intra-hour solar energy predictions

Quentin¹,

Arbod²,

Lasenby³

2022

Preprint

View full text Add to dashboard Cite

Integration of intermittent renewable energy sources into electric grids in large proportions is challenging. A wellestablished approach aimed at addressing this difficulty involves the anticipation of the upcoming energy supply variability to adapt the response of the grid. In solar energy, short-term changes in electricity production caused by occluding clouds can be predicted at different time scales from all-sky cameras (up to 30-min ahead) and satellite observations (up to 6h ahead). In this study, we integrate these two complementary points of view on the cloud cover in a single machine learning framework to improve intra-hour (up to 60-min ahead) irradiance forecasting. Both deterministic and probabilistic predictions are evaluated in different weather conditions (clear-sky, cloudy, overcast) and with different input configurations (sky images, satellite observations and/or past irradiance values). Our results show that the hybrid model benefits predictions in clear-sky conditions and improves longer-term forecasting. This study lays the groundwork for future novel approaches of combining sky images and satellite observations in a single learning framework to advance solar nowcasting.

show abstract

“…Para generar la predicción de producción de irradiancia, se ha desarrollado un modelo de Deep Learning basado en el presentado en [11] y cuya arquitectura fundamental puede ser observada en la Figura 1. Como se puede ver, esta utiliza como entrada principal estimaciones de irradiancia pasada en el área que rodea la localización objetivo.…”

Section: Modelo De Predicción Probabilística De Irradianciaunclassified

Dimensionado de sistema de almacenamiento para hibridación con FV a partir de predicciones probabilísticas de irradiancia

Pérez¹,

Barreda²,

Tamarit³

et al. 2021

Xlii Jornadas De Automática : Libro De Actas

Self Cite

View full text Add to dashboard Cite

En este trabajo se propone una metodología para dimensionar el sistema de almacenamiento de un sistema híbrido con fotovoltaica (FV) para uso doméstico, a partir de predicciones probabilísticas de la producción solar. Se introduce para ello un modelo basado en Deep Learning que, a partir de estimaciones de la irradiancia pasada en el área que rodea la localización objetivo, obtiene las predicciones de distintos percentiles de la producción FV. El dimensionamiento se realiza mediante una optimización lineal que utiliza la función cuantil para garantizar, con un cierto nivel de confianza, que se satisface un perfil de demanda tipo. Finalmente, se introducen y discuten resultados en cuanto a la violación de restricciones que se produce con diferentes tamaños del sistema de almacenamiento mostrándose que, cuando este es superior a 3 h a potencia nominal de la instalación FV, las restricciones se satisfacen en más del 99% de las ocasiones.

show abstract

A deep learning model for intra-day forecasting of solar irradiance using satellite-based estimations in the vicinity of a PV power plant

Cited by 43 publications

References 26 publications

Influence of the Intraday Electricity Market Structure on the Degradation of Li‐Ion Batteries Used to Firm Photovoltaic Production

Influence of the Intraday Electricity Market Structure on the Degradation of Li‐Ion Batteries Used to Firm Photovoltaic Production

Omnivision forecasting: combining satellite observations with sky images for improved intra-hour solar energy predictions

Dimensionado de sistema de almacenamiento para hibridación con FV a partir de predicciones probabilísticas de irradiancia

Contact Info

Product

Resources

About