scholarly journals An LSTM Short-Term Solar Irradiance Forecasting Under Complicated Weather Conditions

IEEE Access ◽  
2019 ◽  
Vol 7 ◽  
pp. 145651-145666 ◽  
Author(s):  
Yunjun Yu ◽  
Junfei Cao ◽  
Jianyong Zhu
Keyword(s):  
Solar Irradiance ◽  
Weather Conditions ◽  
Short Term

Comparison of short-term solar irradiance forecasting methods when weather conditions are complicated

2018 ◽  
Vol 10 (5) ◽  
pp. 053501 ◽  
Author(s):  
Yunjun Yu ◽  
Junfei Cao ◽  
Xiaofeng Wan ◽  
Fanpeng Zeng ◽  
Jianbo Xin ◽  
...  
Keyword(s):  
Solar Irradiance ◽  
Weather Conditions ◽  
Short Term ◽  
Forecasting Methods

scholarly journals Performance Evaluation of Neural Network-Based Short-Term Solar Irradiation Forecasts

Energies ◽  
2021 ◽  
Vol 14 (11) ◽  
pp. 3030
Author(s):  
Simon Liebermann ◽  
Jung-Sup Um ◽  
YoungSeok Hwang ◽  
Stephan Schlüter
Keyword(s):  
Neural Network ◽  
Goodness Of Fit ◽  
Renewable Energy Sources ◽  
Weather Conditions ◽  
Solar Irradiation ◽  
Weather Data ◽  
Short Term ◽  
Climate Conditions ◽  
Lstm Network ◽  
The Time Domain

Due to the globally increasing share of renewable energy sources like wind and solar power, precise forecasts for weather data are becoming more and more important. To compute such forecasts numerous authors apply neural networks (NN), whereby models became ever more complex recently. Using solar irradiation as an example, we verify if this additional complexity is required in terms of forecasting precision. Different NN models, namely the long-short term (LSTM) neural network, a convolutional neural network (CNN), and combinations of both are benchmarked against each other. The naive forecast is included as a baseline. Various locations across Europe are tested to analyze the models’ performance under different climate conditions. Forecasts up to 24 h in advance are generated and compared using different goodness of fit (GoF) measures. Besides, errors are analyzed in the time domain. As expected, the error of all models increases with rising forecasting horizon. Over all test stations it shows that combining an LSTM network with a CNN yields the best performance. However, regarding the chosen GoF measures, differences to the alternative approaches are fairly small. The hybrid model’s advantage lies not in the improved GoF but in its versatility: contrary to an LSTM or a CNN, it produces good results under all tested weather conditions.

scholarly journals Stellar activity, differential rotation, and exoplanets

2010 ◽  
Vol 6 (S273) ◽  
pp. 89-95 ◽  
Author(s):  
A. F. Lanza
Keyword(s):  
Time Series ◽  
Differential Rotation ◽  
Solar Irradiance ◽  
Giant Planet ◽  
Total Solar Irradiance ◽  
Short Term ◽  
Stellar Activity ◽  
Transiting Planets ◽  
Spot Model ◽  
Spot Activity

AbstractThe photospheric spot activity of some of the stars with transiting planets discovered by the CoRoT space experiment is reviewed. Their out-of-transit light modulations are fitted by a spot model previously tested with the total solar irradiance variations. This approach allows us to study the longitude distribution of the spotted area and its variations versus time during the five months of a typical CoRoT time series. The migration of the spots in longitude provides a lower limit for the surface differential rotation, while the variation of the total spotted area can be used to search for short-term cycles akin the solar Rieger cycles. The possible impact of a close-in giant planet on stellar activity is also discussed.

Short-term scheduling problem in smart grid considering reliability improvement in bad weather conditions

2017 ◽  
Vol 11 (10) ◽  
pp. 2521-2533 ◽  
Author(s):  
Babak Yousefi-khangah ◽  
Saeid Ghassemzadeh ◽  
Seyed Hossein Hosseini ◽  
Behnam Mohammadi-Ivatloo
Keyword(s):  
Smart Grid ◽  
Weather Conditions ◽  
Scheduling Problem ◽  
Short Term ◽  
Reliability Improvement ◽  
Bad Weather

The role of short-term weather conditions in temporal dynamics of fire regime features in mainland Spain

2019 ◽  
Vol 241 ◽  
pp. 575-586 ◽  
Author(s):  
Adrián Jiménez-Ruano ◽  
Marcos Rodrigues Mimbrero ◽  
W. Matt Jolly ◽  
Juan de la Riva Fernández
Keyword(s):  
Temporal Dynamics ◽  
Fire Regime ◽  
Weather Conditions ◽  
Short Term

scholarly journals Short-Term Forecasting of Surface Solar Irradiance Based on Meteosat-SEVIRI Data Using a Nighttime Cloud Index

2015 ◽  
Vol 7 (7) ◽  
pp. 9070-9090 ◽  
Author(s):  
Annette Hammer ◽  
Jan Kühnert ◽  
Kailash Weinreich ◽  
Elke Lorenz
Keyword(s):  
Solar Irradiance ◽  
Short Term ◽  
Short Term Forecasting

scholarly journals Forecasting of solar irradiance and ramp events with all-sky imagers

2021 ◽  
Author(s):  
Stavros-Andreas Logothetis ◽  
Vasileios Salamalikis ◽  
Stefan Wilbert ◽  
Jan Remund ◽  
Luis Zarzalejo ◽  
...  
Keyword(s):  
Solar Irradiance ◽  
Large Scale ◽  
Forecast Error ◽  
Forecast Accuracy ◽  
Short Term ◽  
Event Prediction ◽  
High Penetration ◽  
Spanish Research ◽  
Short Term Forecasting ◽  
Solar Resource

<p>Cloud cameras (all sky imagers/ASIs) can be used for short-term (next 20 min) forecasts of solar irradiance. For this reason, several experimental and operational solutions emerged in the last decade with different approaches in terms of instrument types and forecast algorithms. Moreover, few commercial and semi-prototype systems are already available or being investigated. So far, the uncertainty of the predictions cannot be fully compared, as previously published tests were carried out during different periods and at different locations. In this study, the results from a benchmark exercise are presented in order to qualify the current ASI-based short-term forecasting solutions and examine their accuracy. This first comparative measurement campaign carried out as part of the IEA PVPS Task 16 (https://iea-pvps.org/research-tasks/solar-resource-for-high-penetration-and-large-scale-applications/). A 3-month observation campaign (from August to December 2019) took place at Plataforma Solar de Almeria of the Spanish research center CIEMAT including five different ASI systems and a network of high-quality measurements of solar irradiance and other atmospheric parameters. Forecasted time-series of global horizontal irradiance are compared with ground-based measurements and two persistence models to identify strengths and weaknesses of each approach and define best practices of ASI-based forecasts. The statistical analysis is divided into seven cloud classes to interpret the different cloud type effect on ASIs forecast accuracy. For every cloud cluster, at least three ASIs outperform persistence models, in terms of forecast error, highlighting their performance capabilities. The feasibility of ASIs on ramp event detection is also investigated, applying different approaches of ramp event prediction. The revealed findings are promising in terms of overall performance of ASIs as well as their forecasting capabilities in ramp detection.  </p>

Sign in / Sign up

Export Citation Format

Share Document