committable generator constraint fails when min_up_time exceeds length of snapshots

[camiloromers]

Checklist

• I am using the current master branch or the latest release. Please indicate:
• test whether potential bug persists for more than the one snapshot @camiloromers.

Describe the Bug

When I try to run a OPF using the network folder attached below I got the following error:

ERROR: Rule failed when generating expression for constraint
    gen_up_time_force_0 with index 1: IndexError: index 1 is out of bounds for
    axis 0 with size 1
ERROR: Constructing component 'gen_up_time_force_0' from data=None failed:
    IndexError: index 1 is out of bounds for axis 0 with size 1

May I know what's missing in my network or any clue relating the error I am getting?

Thanks for your help.

[L2RPN_2020_ieee_118_pypsa.zip](https://github.com/PyPSA/PyPSA/files/4071575/L2RPN_2020_ieee_118_pypsa.zip)

[FabianHofmann]

Hey, I would not directly claim it a bug.

The error comes from having a network with one snapshot while ramp limits are given. This collides in the lopf. I think it makes sense to make an assertion for this case. Can you please test it for more than the one snapshot?

[camiloromers]

Thanks for your help. I will test it.

One adicional question please. When you run a linear OPF given n snapshots, you are running n independent optimizers or it is just one in the whole space of the snapshots?

[fneum]

network.lopf() will formulate a single optimisation problem including all snapshots of the network to include time-coupled constraints due to storage units or stores. If you want to solve an operational problem (i.e. no capacity expansion) and do not have storage units or stores, you can also run the optimisation for each snapshot individually. Pass the relevant snapshot via network.lopf(snapshots=network.snapshots[0]).

[camiloromers]

Thanks for your feedback. I tested my grid with more than one snapshots but I am having same issue. Any other clue?

[FabianHofmann]

weird... can you send the network (with multiple snapshots now) again?

[camiloromers]

Here you have it again.. Am I doing something wrong when I model my grid? .... Thanks for your support.

ieee118_pypsa.zip

[camiloromers]

I was running some other tests and I found the variable which is causing problems are min_up_time and min_down_time. How they are linked with up_time_before and down_time_before? Apparently something is missing in the last variables when `min_up/down_time are set up.

[FabianHofmann]

Actually, the min_up_time for some of your generators is longer than the number of snapshots:

gen_72_31     576
gen_64_26    2016
gen_89_40    2016

This fails at a specific point in the assignments of the constraints. Can you clip your min_up_time to the length of snapshots. Like this

n.generators.min_up_time.clip(upper=len(n.snapshots), inplace=True)

[camiloromers]

It fix my problem... Thanks a lot

[fneum]

Keeping it open as this case should be caught in the code. Cf. ongoing pull request #128.