For comparison I mostly refer to electric car numbers which do exist for a wide variety of cars (including cars in use by customers) and are all more or less the same so I think we can trust them.
A few diaries ago I said I would ask http://www.effet-de-serre.gouv.fr/ and they just answered me today citing page 33 of this PDF:
http://www.effet-de-serre.gouv.fr/images/documents/memento.pdf
Which gives the following relative energy use for passenger-km:
Walk/bike 0 Paris Tramways 0.10 Paris subway 0.13 Paris RER 0.14 Paris "banlieue" trains 0.21 Paris area buses 0.44 France buses 0.56 Motorcycle 0.62 Individual car 1.0
(I don't know if the car data is for one passenger only ... these "green" paper could be more informative).
So there is a Citadis model that does power injection
The way I understand it, it an inherent feature (it should be with IGBT inverters), and use depends on whether the tram line power supply was made suited. With some more search, I found other examples of active use (Rotterdam, Sydney, Basel, etc.), so it may not be uncommon. Recovery rates are usually low, but in Basel, a phenomenal 41.6% was achieved with Siemens Combinos in a test. The latter link also gives you kWh/km figures (page 8). *Lunatic*, n. One whose delusions are out of fashion.
If I do the math on the Basel data:
