Recent fever changes are not picked up by heat_network_mt and heat_network_lt

[mabijkerk]

On production, if I open a blank nl2019 scenario and set the households space heating technologies to 50% for MT heat networks, I immediately get a backup gas burner to supply the heat demand:

However, when I do the same on master, no bakcup gas burner is used:

This is not due to large deficits. Instead, something seems to go wrong on the backup burner node:

More specifically, none of the dispatchable heat source on the MT heat network seem to be working. This is the same for the LT heat network. HT heat network seems to be working fine.

[noracato]

First breakthrough:
If in the node households_final_demand_mt_steam_hot_water.final_demand you set the demand to e.g. 15.0 instead of 0.0, then all future calculations start working again. So it looks like some calculations don't happen in the future when the present is 0, I'll continue looking!

[mabijkerk]

Heat networks on MT and LT now are closer to what we would expect:

The only thing that now does not seem to be working is the heat provided to account for distribution losses in the MT heat network:

More is demanded from the backup burner in steam_hot_water than what is actually used in network_gas. The difference seems to be in the energy_heat_distribution_mt_loss.

[noracato]

Done

[mabijkerk]

LGTM!