DCOPF Verification¶

Conclusion¶

For test cases, DCOPF results from AMS are identical to that from PYPOWER.

NOTE: This verification works with PYPOWER *v5.1.16, where v5.1.17* yields unexpected results.

In [1]:

import datetime

import numpy as np
import pandas as pd

import ams
import andes

import pypower.api as pyp

In [2]:

print("Last run time:", datetime.datetime.now().strftime("%Y-%m-%d %H:%M:%S"))

print(f'ams: {ams.__version__}')

Last run time: 2024-11-24 21:03:09
ams: 0.9.12

In [3]:

ams.config_logger(stream_level=30)

Generator Output¶

Using built-in MATPOWER cases as inputs.

In [4]:

cases = [
    ams.get_case('matpower/case14.m'),
    ams.get_case('matpower/case39.m'),
    ams.get_case('matpower/case118.m'),
    ams.get_case('npcc/npcc.m'),
    ams.get_case('wecc/wecc.m'),
    ams.get_case('matpower/case300.m'),]

case_names = [case.split('/')[-1].split('.')[0] for case in cases]

In [5]:

ams_obj = np.zeros(len(cases))
pyp_obj = np.zeros(len(cases))

for i, case in enumerate(cases):
    sp = ams.load(case, setup=True)
    sp.DCOPF.init()
    sp.DCOPF.run(solver='CLARABEL')
    ams_obj[i] = sp.DCOPF.obj.v

    ppc = ams.io.pypower.system2ppc(sp)
    ppopt = pyp.ppoption(VERBOSE=0, OUT_ALL=0, PF_ALG=1, OPF_ALG_DC=200)
    ppc_sol = pyp.rundcopf(ppc, ppopt)
    pyp_obj[i] = ppc_sol['f']

Building system matrices
Parsing OModel for <DCOPF>
Evaluating OModel for <DCOPF>
Finalizing OModel for <DCOPF>
<DCOPF> solved as optimal in 0.0065 seconds, converged in 11 iterations with CLARABEL.
Building system matrices
Parsing OModel for <DCOPF>
Evaluating OModel for <DCOPF>
Finalizing OModel for <DCOPF>
<DCOPF> solved as optimal in 0.0077 seconds, converged in 8 iterations with CLARABEL.
Building system matrices
Parsing OModel for <DCOPF>
Evaluating OModel for <DCOPF>
Finalizing OModel for <DCOPF>
<DCOPF> solved as optimal in 0.0144 seconds, converged in 11 iterations with CLARABEL.
Building system matrices
Parsing OModel for <DCOPF>
Evaluating OModel for <DCOPF>
Finalizing OModel for <DCOPF>
<DCOPF> solved as optimal in 0.0164 seconds, converged in 12 iterations with CLARABEL.
Building system matrices
Parsing OModel for <DCOPF>
Evaluating OModel for <DCOPF>
Finalizing OModel for <DCOPF>
<DCOPF> solved as optimal in 0.0160 seconds, converged in 12 iterations with CLARABEL.
Building system matrices
Parsing OModel for <DCOPF>
Evaluating OModel for <DCOPF>
Finalizing OModel for <DCOPF>
<DCOPF> solved as optimal in 0.0226 seconds, converged in 13 iterations with CLARABEL.

In [6]:

np.allclose(ams_obj, pyp_obj, atol=1e-6)

Out[6]:

True

LMP¶

In [7]:

sp2 = ams.load(ams.get_case('pglib/pglib_opf_case39_epri__api.m'),
               setup=True,
               no_output=True,
               default_config=True)

In [8]:

sp2.DCOPF.run(solver='CLARABEL')

Building system matrices
Parsing OModel for <DCOPF>
Evaluating OModel for <DCOPF>
Finalizing OModel for <DCOPF>
<DCOPF> solved as optimal in 0.0072 seconds, converged in 13 iterations with CLARABEL.

Out[8]:

True

In [9]:

ppc2 = ams.io.pypower.system2ppc(sp2)
ppc2_sol = pyp.rundcopf(ppc2, ppopt)

Nodal price

In [10]:

np.allclose(sp2.DCOPF.pi.v / sp2.config.mva,
            ppc2_sol['bus'][:, 13],
            atol=1e-6)

Out[10]:

True

Bus angle

In [11]:

np.allclose(sp2.DCOPF.aBus.v * andes.shared.rad2deg,
            ppc2_sol['bus'][:, 8],
            atol=1e-6)

Out[11]:

True

In [12]:

pd.DataFrame({'ams_LMP': sp2.DCOPF.pi.v / sp2.config.mva,
              'pyp_LMP': ppc2_sol['bus'][:, 13],
              'ams_aBus': sp2.DCOPF.aBus.v * andes.shared.rad2deg,
              'pyp_aBus': ppc2_sol['bus'][:, 8]}).round(4)

Out[12]:

	ams_LMP	pyp_LMP	ams_aBus	pyp_aBus
0	29.9988	29.9988	-25.6184	-25.6184
1	34.2147	34.2147	-26.8519	-26.8519
2	35.6012	35.6012	-30.4965	-30.4965
3	38.1542	38.1542	-28.8548	-28.8548
4	20.1311	20.1311	-24.4545	-24.4545
5	20.9823	20.9823	-22.9110	-22.9110
6	21.0918	21.0918	-26.4821	-26.4821
7	21.1466	21.1466	-27.2531	-27.2531
8	24.8701	24.8701	-24.5329	-24.5329
9	24.8047	24.8047	-21.2851	-21.2851
10	23.5694	23.5694	-21.8833	-21.8833
11	24.8047	24.8047	-22.4869	-22.4869
12	26.0400	26.0400	-22.7386	-22.7386
13	34.3071	34.3071	-26.2107	-26.2107
14	34.6822	34.6822	-29.2227	-29.2227
15	34.8446	34.8446	-27.6899	-27.6899
16	34.9985	34.9985	-29.5901	-29.5901
17	35.2283	35.2283	-30.6918	-30.6918
18	34.8446	34.8446	-21.3650	-21.3650
19	34.8446	34.8446	-23.3981	-23.3981
20	34.8446	34.8446	-24.2513	-24.2513
21	34.8446	34.8446	-17.0667	-17.0667
22	34.8446	34.8446	-17.0905	-17.0905
23	34.8446	34.8446	-27.0536	-27.0536
24	34.3071	34.3071	-25.0737	-25.0737
25	34.6544	34.6544	-28.2256	-28.2256
26	34.8125	34.8125	-30.9591	-30.9591
27	34.6544	34.6544	-24.1366	-24.1366
28	34.6544	34.6544	-19.8996	-19.8996
29	34.2147	34.2147	-26.7668	-26.7668
30	20.9823	20.9823	0.0000	0.0000
31	24.8047	24.8047	-11.0741	-11.0741
32	34.8446	34.8446	-14.3249	-14.3249
33	34.8446	34.8446	-14.2720	-14.2720
34	32.3065	32.3065	-9.5084	-9.5084
35	18.1575	18.1575	-3.0644	-3.0644
36	31.5502	31.5502	-12.8112	-12.8112
37	34.6544	34.6544	-9.0065	-9.0065
38	27.4344	27.4344	-22.5664	-22.5664