data: update hvdc and cost curve according to real values

[BainanXia]

Purpose

According to the results of Scenario 462, we finalize the fixes on the HVDC lines between interconnections and the scalings of cost curve parameters in order to align the LMPs with historical values. We apply following scale parameters on the cost curves for each interconnection:

western_scale_factor = 1.34
texas_scale_factor = 1.09
eastern_scale_factor = 1.67

What the code is doing

No code changes, only the data in dcline.csv and gencost.csv

Time to review

10-15 mins

[danielolsen]

DC lines look good.

Comparing develop to hvdc_cost_curve_update1, I see the following range of scale factors:
c2: 1.09 for all Texas gens, 1.34 for all Western gens, 1-1.67 for Eastern gens.
c1: 1-1.09 for Texas gens, 1-1.34 for Western Gens, 1-1.67 for Eastern gens
c0: 1-1.09 for Texas gens, 1-1.34 for Western Gens, 1-1.67 for Eastern gens

This is via

>>> import pandas as pd
>>> gencost_ref = pd.read_csv('gencost_develop.csv', index_col=0)
>>> gencost_test = pd.read_csv('gencost_test.csv', index_col=0)
>>> c2_scaling = gencost_test.c2 / gencost_ref.c2
>>> c1_scaling = gencost_test.c1 / gencost_ref.c1
>>> c0_scaling = gencost_test.c0 / gencost_ref.c0
>>> c2_scaling.groupby(gencost_ref.interconnect).min()
interconnect
Eastern    0.999998
Texas      1.090000
Western    1.340000
Name: c2, dtype: float64
>>> c2_scaling.groupby(gencost_ref.interconnect).max()
interconnect
Eastern    1.670008
Texas      1.090000
Western    1.340000
Name: c2, dtype: float64
...

Some generator parameters aren't being scaled? Or is there an issue with how I'm looking into the numbers?

[BainanXia]

@danielolsen Only coal, ng and dfo generators are scaled accordingly.

[danielolsen]

I find the same thing when I groupby interconnect and plant.type. Those plants are scaled properly, across all interconnects, nuclear is 1, and renewables are all NaN.

[BainanXia]

@danielolsen That was the plan, right?

[danielolsen]

@BainanXia It checks out. Nuclear should almost never be setting LMPs, so should not influence HVDC powerflow.

[BainanXia]

@danielolsen So is it good to go or we should scale nuclear as well?

[danielolsen]

It's good to go!