How to forecast green energy production
Franck Freycenon – Head of E&U Solutions, Atos
One of the main challenges holding back the growth of renewable energy sources such as solar PV and wind power is their inherent variability. Effective management of the energy grid requires a constant balance between electricity production and consumption. So, how can forecasters and energy specialists manage to predict green energy production?
Knowing how to balance the energy mix
The variability of renewables has a direct financial impact along the whole value chain. It affects the energy producers, who are constantly at risk of either over-producing (and having to somehow get rid of the surplus so as not to overload distribution infrastructures) or under-producing (and losing potential revenues). So producers have to factor that uncertainty into their financial and operational models.
This affects the aggregators, who buy electricity from smaller producers to resell wholesale and play a key role in this balancing process on the networks. It also affects distributors, who need a highly accurate and up-to-date picture of current production in order to offer a high-quality service and that’s even before you take into account so-called ‘prosumers’, customers who actually generate some of their own electricity.
Being able to forecast production accurately is therefore crucial for the economic viability of the renewables sector, as well as for its ability to compete with other sources of energy and contribute to the fight against climate change. The European Union, most notably, has set itself the target for renewables to provide 32% of end-user energy consumption by 2030 (compared with 18.9% in 2018).
Putting the theory into practice
Forecasting energy production is a complex process. The French national meteorological service, Météo-France, has taken a major step forward by developing models for both wind and solar generation. The models – known as WF Wind and MF Solar – facilitates simulations that can be used to assess a potential energy production site. However, the energy produced by a wind turbine is directly proportional to the wind-speed cubed, so errors very quickly get amplified.
If you’re looking for more accurate forecasts, you have to go much further than these theoretical models will allow. Here is where machine learning comes in to play. By comparing historical production data with current measurements, taken moment by moment, an algorithm can build an extremely precise forecasting model which is unique to each particular installation.
Learning to correct
Of course, as with any approach based around data, everything depends on the quality of that data. If there is a problem with one element (perhaps a single solar panel), this needs to be removed from the dataset. That’s why the models that Atos has designed adapt and can relearn from cleaned data if it has detected a divergence in the trend.
It is by combining high-quality weather forecasts, proven physical models and expertise in algorithms that we can take out the uncertainty attached to renewable energy. This can not only provide direct assistance to those managing renewable energy networks, but can also lessen many of the barriers to further investment and expansion. By providing energy providers with greater certainty in a more timely way, these data systems can play a key role in driving the future growth of renewables.