Capabilities of short-term forecasting of solar energy
The share of solar energy is growing rapidly in power around the world. Solar variability can affect the operation of power systems. Achieving predictability of solar energy allows for a better balance of electricity consumption. There are various resources to predict solar and photovoltaic energy, including traditional measuring weather data, information of active solar power plants, aerospace surveillance data on cloud cover, various analytical models (Numerical Weather Prediction).
Short-term forecasting includes time intervals for minutes and hours; it is necessary for frequency control and load balancing.
Mid-term forecast, from several hours to several days, based on market requirements for energy trade. Long-term forecasting is necessary for grid development planning and economic analysis,
and performed in seasonal and annual time horizons.
The physical approach to predicting deal with solar and photovoltaic energy behaviours, and statistical approach based primarily on historical data to identify trends. If the forecast concerning a large number of areas, some objects are modelled and methods of extrapolation or interpolation are used.
Images of the sky are using with the methods of tracking the movement of clouds in the sky photographs. For satellite imagery, a similar approach is used for a longer time horizon because of the spatial and temporal resolution. For deterministic approach a certain level of solar energy is predicted, the stochastic indicates an additional level of uncertainty. The combination of techniques provides better accuracy of forecasts. To assess the accuracy of prediction the average deviation, mean square error, mean absolute error, standard deviation are used. The accuracy of the forecast is affected by the local climate, the amount of area or number of areas, forecasting horizon, a precision of the measuring equipment. The emergence of intelligent networks and power systems management forms its own requirements for predictability and encourages new developments in forecasting.
2. Photovoltaic and Solar Forecasting: State of the Art. Report IEA. Electronic resource: www.iea-pvps.org.
3. A.Haevskyy, O.Melnyk, PV Power Forecasting based on satellite data. Materials of XIV International scientific conference "Renewable Energy of XXI Century ", 2013. – P. 248–249.
4. A.Lorenzo, W.Holmgren at al. Intra-hour solar power forecasts using a real-time irradiance monitoring network.
Електронний ресурс: https://ams.confex.com/ams/...Paper267978.html
5. M.Kuznietsov Possibilities of short-term forecasting of wind speed on wind farm // Kyiv Ukrainian: Renewable Energy. – 2010. – № 4. – P. 40–47.
6. A.Bracale at al. A Bayesian Method for Short-Term Probabilistic Forecasting of Photovoltaic Generation in Smart Grid Operation and Control // Energies – 2013, №6. – Р.733–747.
Abstract views: 60 PDF Downloads: 211
This work is licensed under a Creative Commons Attribution 4.0 International License.