Standby-cost the wrong way round?

[Cellophil]

Hi,

Standby cost is a newish feature that according to the docs

  | Stand-by cost for operating the generator at null power output. |

The code on the other hand is:

status = n.model.variables[f"{c}-status"].loc[:, stand_by_cost.columns]
objective.append((status * stand_by_cost).sum())

which seems to give cost to the Generator once it's running. Unless the stand_by_cost gets a sign flip but it doesn't look like it. I also tested a toy model - with a switched off Generator the objective function remains zero.

m = pypsa.Network()
m.add('Bus', 'mybus')
m.add('Generator', bus='mybus', p_nom=100, stand_by_cost=100, name='mygen', committable=True, marginal_cost=1)
m.set_snapshots(np.arange(10))
m.add('Load', bus='mybus', name='myload')
m.loads_t['p_set'].loc[:,'myload'] = np.zeros(10)
m.loads_t['p_set'].iloc[0,0] = 1
m.optimize()

gives me a objective function of 101; 100 for one snapshot of running plus 1 fuel cost.

I would actually prefer the current code version, offset cost for running a unit is very useful. It immediately acts as an efficiency curve. On the other hand "standby" implies the Generator is switched off, that could confuse users.
Also, standby cost does not seem to show up in the statistics module yet.

[Cellophil]

Now that I think about it it could be that I misinterpreted the doc, maybe "operating generator with null power" refers to state=on [...regardless of power output]. Then it would be just a matter of maybe adding a clarifying sentence to the doc.