Solar power in the Netherlands: is that a duck? (Updated)

This post has been originally published in July 2020. This is an updated version with the data available at the beginning of 2024

Although the Netherlands is a not a country that everyone would define “sunny”, solar power is becoming there more and more common. According to the ENTSO-E Transparency Platform there are now 22590 MW of installed solar power (this number was 7226 in 2020), slightly less than 1300 Watt of solar power per person living in the country. This makes the Netherlands the first country in the world for solar watt per capita (last time I checked Australia had about 1100 Watt per capita).

Only ten years ago, solar power in the Netherlands was almost non-existent with barely 750 MW (45 Watt per person!). Can we observe the effects of this rapid penetration of solar power in the national electricity demand data?

We can access again the data from the ENTSO-E Transparency Platform, which provides the demand observed by the national transmission system operator (TSO). Given that almost all the photovoltaic plants are installed in the distribution grid or behind-the-meter (rooftop), the generation of solar power is seen by the TSO as a “negative” demand.

Then the question: can we observe a duck? The “duck curve” is the term invented by the Californian system operator to describe the impact of solar generation on the observed electricity demand. The Wikipedia page is quite nice actually showing also a couple of nice graphs, with ducks of course.

We plot now the average daily curve per month - only for the working days - from January 2016 to December 2023.

Is that a duck? Well, the “duckification” of the daily curve is complete. Now the curve is more similar to a canyon. However, we can try to check another thing.

Firstly, let’s check what was the trend of the hourly demand in the last years. We would like to see two things:

1. The average hourly demand
2. The average hourly demand only between 9:00 and 16:00, the hours where the back of the duck should be more visible.

We apply a rolling mean of 365 days.

Apparently the demand was more or less stable during the last years (except for the drop due to the COVID-19 measures and to the increase of energy prices in the recent year) but instead the demand between 9:00 and 16:00 dropped visibly, especially in the last year.

Another step: let’s analyse the variability. The more the solar power capacity and the more we expect to see a day-by-day variability during the day (the weather in the Netherlands can change very quickly). We want to visualise the “gap” between two consecutive (working) days: we expect to see larger gaps with more solar capacity. We are still analysing the demand during the period 9:00-16:00.

Then the average difference of demand between two consecutive days is increasing from 500 MW in 2017-2018 to more than 1000 during 2022-2023.

According to Solar Market Europe, the Netherlands might add 25 GW of solar capacity by 2027. It will be interesting then to see how the charts shown in this page will evolve and change.