3 ms·
French grid data from RTE is available from http://www.gridwatch.templar.co.uk/france/ http://www.gridwatch.templar.co.uk/france/ I took the frequency data fro
by mhandley 9y ago
French grid data from RTE is available from http://www.gridwatch.templar.co.uk/france/ http://www.gridwatch.templar.co.uk/france/
I took the frequency data from the last 1200 days, and calculated the cumulative drift. The raw drift graph, assuming 50Hz is nominal, is here:
http://nrg.cs.ucl.ac.uk/mjh/drift-raw.png http://nrg.cs.ucl.ac.uk/mjh/drift-raw.png
This seems bogus; there's a constant drift.
If I compensate for that drift by assuming the nominal grid frequency is 49.9972Hz, the data looks a little more sane.
http://nrg.cs.ucl.ac.uk/mjh/drift.png http://nrg.cs.ucl.ac.uk/mjh/drift.png
You can clearly see the 6 minute deviation this year - it's pretty striking, and certainly nothing this drastic has happened in the last three years. But what's with the positive offset over the previous year? Not entirely sure how much I trust the RTE data.
- Aissen 9y agoAwesome work, thanks!
- ajnin 9y agoFor what it's worth, I have an alarm clock connected to the French grid, and until last month it's been perfectly on time, so assuming it is tuned to 50Hz there is probably something wrong with the source data.
- mhandley 9y agoAgreed - without knowing more about the data provenance, all I can show here are deviations from the long-term trend. It's enough to see the recent problem clearly, but I wouldn't trust the increase after day 900 to be real. Most likely something changed in how they report the data around that time.