Solar power forecasting


 * Nominated for deletion at Wikipedia:Wikipedia:Articles for deletion/Solar power forecasting

Photovoltaic Power Production is increasing nowadays. Forecast information is essential for an efficient use, the management of the electricity grid and for solar energy trading.

Solar Photovoltaic systems transform solar energy into electric power. The power output depends on the incoming radiation and on the solar panel characteristics. To achieve an accurate forecast we need to have a deep knowledge of the Sun´s path, the atmosphere´s condition and the scattering processes. We also have to take into account the characteristics of the solar plant.

Generation forecasting
The energy generation forecasting problem is closely linked to the problem of weather variables forecasting. Indeed, this problem is usually split into two parts, on one hand focusing on the forecasting of solar PV or any other meteorological variable and on the other hand estimating the amount of energy that a concrete power plant will produce with the estimated meteorological resource. In general, the way to deal with this difficult problem is usually related to the spatial and temporal scales we are interested in, which yields to different approaches that can be found in the literature. In this sense, it is useful to classify these techniques depending on the forecasting horizon, so it is possible to distinguish between nowcasting (forecasting 3-4 hours ahead), short-term forecasting (up to 7 days ahead) and long-term forecasting (months, years…)

What is nowcasting
Nowcasting comprises the detailed description of the current weather along with forecasts up to 3-4 hours. This very short-term forecasting services are very important for grid operators in order to guarantee the grid stability and for those power plants that can be considered manageable, at least in a certain degree, such as solar thermal power plants. Nowcasting services are usually related to very high temporal resolution (a forecast every 10 or 15 minutes), so automatic weather data acquisition and processing is a major requirement in order to develop these techniques. Several approaches can be found in the literature, which mainly depend on the type of data that is treated to estimate future values of meteorological variables:

a)	Statistical techniques are usually based on time series processing of meteorological measured data, which is used as training data to tune the parameters of a model (I. Espino eta al, 2011). These techniques include the use of any kind of statistical approach, such as autoregressive moving average (ARMA, ARIMA,…), neural networks, support vector machines, etc. These approaches are usually benchmarked to a persistence approach in order to evaluate their improvements. This persistence approach just assumes that any variable at time step t is the value it took in a previous time.

b)	Since the launch of Earth observing satellites, such as MSG, nowcasting techniques have also been developed from an image processing point of view. The main advantage of these techniques is the possibility to monitorize a lot of meteorological information in almost real time. This high value information is used as input to physical models based on image processing techniques that provide an estimation of future atmospheric values, as described in Alvarez et al, 2010.

Solar PV short-term forecasting
Short-term forecasting provides predictions up to 7 days ahead. This kind of forecast is also valuable for grid operators in order to make decisions of grid operation, as well as, for electric market operators. Under this perspective, the meteorological resources are estimated at a different temporal and spatial resolution. This implies that meteorological variables and phenomena are looked from a more general perspective, not as local as nowcasting services do. In this sense, most of the approaches make use of different numerical weather prediction models (NWP) that provide an initial estimation of weather variables. Currently, several models are available for this purpose, such as Global Forecasting Service (GFS) or data provided by the European Center for Medium Range Weather Forecasting (ECMWF) (Ecmwf web site). These two models are considered the state of the art of global forecast models, which provide meteorological forecasts all over the world. In order to increase spatial and temporal resolution of these models, other models have been developed which are generally called mesoscale models. Among others, HIRLAM, WRF or MM5 are the most representative of these models since they are widely used by different communities. To run these models a wide expertise is needed in order to obtain accurate results, due to the wide variety of parameters that can be configured in the models. In addition, sophisticated techniques such as data assimilation might be used in order to produce more realistic simulations. Finally, some communities argue for the use of post-processing techniques, once the models’ output is obtained, in order to obtain a probabilistic point of view of the accuracy of the output. This is usually done with ensemble techniques that mix different outputs of different models perturbed in strategic meteorological values and finally provide a better estimate of those variables and a degree of uncertainty, like in the model proposed by Bacher et al (2009)

Solar PV long-term forecasting
Long-term forecasting usually refers to forecasting of the annual or monthly available resource. This is useful for energy producers and to negotiate contracts with financial entities or utilities that distribute the generated energy. In general, these long-term forecasting is usually done at a lower scale than any of the two previous approaches. Hence, most of these models are run with mesoscale models fed with reanalysis data as input and whose output is postprocessed with statistical approaches based on measured data.

Energetic Models
Any output from any model described above must then be converted to the electric energy that a particular solar PV plant will produce. This step is usually done with statistical approaches that try to correlate the amount of available resource with the metered power output. The main advantage of these methods is that the meteorological prediction error, which is the main component of the global error, might be reduced taking into account the uncertainty of the prediction. As it was mentioned before and detailed in Heinemann et al, these statistical approaches comprises from ARMA models, neural networks, support vector machines, etc. On the other hand, there also exist theoretical models that describe how a power plant converts the meteorological resource into electric energy, as described in Alonso et al. The main advantage of this type of models is that when they are fitted, they are really accurate, although they are too sensitive to the meteorological prediction error, which is usually amplified by these models. Hybrid models, finally, are a combination of these two models and they seem to be a promising approach that can outperform each of them individually.

Weather prediction models

 * Documentation of HiRLAM at ECMWF
 * Description of HiRLAM at KNMI
 * GFS

Electricity market

 * Power exchange APX (Netherlands)
 * System Operator TenneT (Netherlands)
 * Nord Pool (Scandinavia)
 * Operadora del Mercado Ibérico de Energía - Polo Español, S.A. (Spain)

Solar PV power forecasting methods

 * Enfor(Denmark)
 * gWISE SOLAR - (Gnarum)
 * Solar Plant Generation Forecast (Meteologica)
 * Aeolis(Dutch)
 * Datameteo(Italy)
 * Ecofys (UK)
 * Irradiance Data (Germany)
 * AleaSolar (AleaSoft)

Solar radiation map

 * Solar radiation world map