Storage Unit Behavior and Marginal Prices

[Andy Boomer]

Hi,

I am trying to understand how to best model market participation of grid scale batteries.

A battery operator should be able to both control the price at which it charges and the price at which it discharges.

I have been able to test out that the marginal cost works to control the price at which a battery discharges power, just like other generators, but I am having trouble figuring out how to control this on the charging side.

I am not immediately sure what the best way to implement this would be, but I would expect to be able to specify one or more sets of marginal costs (or bids/offers for a battery, but could be the same theoretically).

- The battery is willing to charge if the resulting bus marginal price is less than or equal to to the submitted cost/bid/offer (and physical constraints are satisfied)

- The battery is willing to discharging if the resulting bus marginal price is greater than or equal to the submitted cost/bid/offer (and physical constraints are satisfied)

I am working with this toy network adapted from one of the docs examples. I can't seem to figure out how to control the load side of the battery. Thanks!

network = pypsa.Network()

network.set_snapshots(range(5))

network.add("Bus", "A")

network.add("Bus", "B")

network.add("Bus", "C")

network.add("Line", "a_b", bus0="A", bus1="B", s_nom=100000, r=0.01, x=0.01)

network.add("Line", "a_c", bus0="A", bus1="C", s_nom=100000, r=0.01, x=0.01)

network.add(

"Load",

"load_a",

bus="A",

p_set=np.array([40000, 40000, 44000, 46000, 50000]),

)

network.add(

"Generator",

"gen_coal",

bus="B",

p_nom=50000,

marginal_cost=30,

p_max_pu=1,

)

network.add(

"Generator",

"gen_solar",

bus="C",

p_nom=40100,

marginal_cost=0,

p_max_pu=[1.0, 1.0, 0.0, 0.0, 0.0]

)

network.add(

"StorageUnit",

f"Pumped hydro",

bus="C",

p_nom=50,

p_min_pu=-1,

p_max_pu=1,

max_hours=6,

marginal_cost=[5, 5, 0, 0, 0],

state_of_charge_initial=0.0,

)

network.optimize()