Atmosphere & Solar Radiation Modeling Group

Desarrollo de un modelo operacional de predicción del recurso solar en escalas de horas a días (SOLCASTING).


Increasing demand for electricity, coupled with a declining fossil fuel base and the consequences of greenhouse gas emissions, have stimulated efforts worldwide for developing cheap and reliable supplies of electricity from renewable sources. Solar energy is probably the only long-term supply-side energy solution that is both large enough and acceptable enough to sustain these planets long term requirements. Power production using solar energy has improved globally and experienced a significant expansion by, primarily, reducing costs and achieving greater conversion efficiency. Spain can be regarded as a reference sit for solar energy applications, since it is one of the world leading countries in electricity production from solar origin sources.

A major challenge for the future will be the solar plants operation management and the integration of the large scale solar yields into existing energy supply structures. The problem arises from the fluctuating character of the solar resource, as compared to conventionally generated electricity, and its dependence on non-deterministic weather patterns. This presents challenges regarding a wide range of aspects of the operation and yield integration of the plants in the electric grid. Particularly, it may lead to harmful thermal stress into the STPP concentrating mirrors or instabilities in the electric network due to transients originated by the clouds movement. As a result, one of the main problems in the operation of STPP is to predict when a cloud will cover the field and how much solar radiation will be attenuated. Therefore, STPPs need forecasts of direct normal solar irradiance (DNI) –the resource they are interested in— for improved operation management and to prevent damages in the facilities. This forecasting, known as nowcasting, has to be provided at a time horizon of 1-2 hours ahead. In addition, as the number of renewable energy plants integrated in the electric grid increases, the management and distribution of the energy in the whole grid becomes more and more complex because of the intermittent availability of solar resource which puts at risk the stability of the supply. This strongly limits the rate of solar facilities that the system is able to hold. In order to maximize their penetration into the distribution grid without outages and congestion, forecasting of solar radiation at least 24 hours ahead is vital.

Therefore, development of forecasts of the solar resources will be certainly of fundamental importance in order to reach the objectives stated by the EU’s member states for renewable energies penetration into the electricity grid by 2020. Particularly, the International Energy Agency (IEA) establishes that “precise local solar forecasts” is the main topic of research interest in the area of the solar energy resources. Both satellite (nowcasting) and NWPs (forecasting) are referenced as the most important tools regarding this area of research. Just as wind forecasting has reached certain level of maturity, solar radiation still need profuse research efforts.

This proposal pursuit the improvement of current methods in forecasting the solar radiation, both global and direct values, at scales from few hours (nowcasting) to few days using NWP models. Accurate information on expected solar irradiance on these time scales are necessary for (i) solar plant operation, (ii) selling the electricity on the spot market with the best possible price and, (iii) allocating the needed balance energy from ancillary sources when no solar energy is available.