Solution & technology
Trading : Risk Decreases with Rising Spatial Spread of Portfolio – Case Study in Italy
December 28th, 2016
We understand intuitively that it is preferable to have a portfolio of solar plants spread over a wide territory. Indeed, thanks to the large-scale statistical effect of production and deviation forecasting, balancing risk and costs are minimized.
Here are the outputs and forecasts for 4 Italian plants.
Plants nr 2, 3 and 4 are respectively at a distance of 2, 10 and 70 km from plant nr 1.
The 3 selected days are characterized by unstable weather conditions and a fairly significant variability.
This variability can however be fully visualized and analyzed as the measurement data are at a time step of 30 minutes. Thus the rapid fluctuations do not appear.
Figure 1 : D + 1 forecasting for 3 consecutive days – plant nr 1
Figure 2 : D + 1 forecasting for 3 consecutive days – plant nr 2, 20 km away from plant nr 1
Figure 3 : D + 1 forecasting for 3 consecutive days – plant nr 3, 10 km away from plant nr 1
The 3 first plants are very close, they are globaly exposed to the same weather conditions and therefore to similar variations, providing thereby little possibility for the large-scale mitigation factor. This is why it is so important for traders to have solar plants in their portfolios that are located in areas subject to different weather conditions.
As plant nr 4 is 70 km far from plant nr 1, it is more likely to be the case with this plant.
December 28th, 2016 is a good example : the production of plant nr 4 is at a high level throughout the day (clear sky) whereas the 3 other plants are subject to severe disturbance in the afternoon leading to a very sudden and sharp drop of production.
Figure 4 : D + 1 forecasting for 3 consecutive days – plant nr 4, 70 km away from plant nr 1