Multi-period Dispatch Simulation

Multi-period dispatch economic dispatch (ED) and unit commitment (UC) is also available.

In this case, we will show a 24-hour ED simulation.

import ams

import datetime

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

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

Last run time: 2024-04-21 17:30:15
ams:0.9.6

ams.config_logger(stream_level=20)

Load Case

sp = ams.load(ams.get_case('5bus/pjm5bus_demo.xlsx'),
              setup=True,
              no_output=True,)

Parsing input file "/Users/jinningwang/Documents/work/mambaforge/envs/amsre/lib/python3.9/site-packages/ams/cases/5bus/pjm5bus_demo.xlsx"...
Input file parsed in 0.0926 seconds.
Zero line rates detacted in rate_b, rate_c, adjusted to 999.
System set up in 0.0025 seconds.

Reginonal Design

The disaptch models in AMS has develoepd with regional structure, and it can be inspected in device Region.

sp.Region.as_df()

	idx	u	name
uid
0	ZONE_1	1.0	ZONE 1
1	ZONE_2	1.0	ZONE 2

In device Bus, the Param zone indicates the zone of the bus. Correspondingly, the region of generator and load are determined by the bus they connected.

sp.Bus.as_df()

	idx	u	name	Vn	vmax	vmin	v0	a0	xcoord	ycoord	area	zone	owner
uid
0	Bus_1	1.0	A	230.0	1.1	0.9	1.0	0.0	0	0	1	ZONE_1	None
1	Bus_2	1.0	B	230.0	1.1	0.9	1.0	0.0	0	0	1	ZONE_1	None
2	Bus_3	1.0	C	230.0	1.1	0.9	1.0	0.0	0	0	2	ZONE_1	None
3	Bus_4	1.0	D	230.0	1.1	0.9	1.0	0.0	0	0	2	ZONE_1	None
4	Bus_5	1.0	E	230.0	1.1	0.9	1.0	0.0	0	0	3	ZONE_1	None

Multi-period Dispatch Base

In AMS, multi-period dispatch involves devices in group Horizon. This group is developed to provide time-series data for multi-period dispatch.

sp.Horizon.models

OrderedDict([('TimeSlot', TimeSlot (0 devices) at 0x2905f8af0),
             ('EDTSlot', EDTSlot (6 devices) at 0x2906045b0),
             ('UCTSlot', UCTSlot (6 devices) at 0x2906049d0)])

We can get the idx of StaticGens.

sp.StaticGen.get_idx()

['Slack_4', 'PV_1', 'PV_3', 'PV_5']

In EDTSlot, Param sd refers the load factors of each region in each time slot, and Param ug represents the generator commitment status in each time slot.

To be more specific, EDT1 has sd=0.0793,0.0, which means the load factor of region 1 is 0.0793 in the first time slot, and 0.0 in the second time slot.

Next, EDT1 has ug=1,1,1,1, and it means the commitment status of generator PV_1, PV_3, PV_5, and Slack_4 are all online.

sp.EDTSlot.as_df()

	idx	u	name	sd	ug
uid
0	EDT1	1.0	EDT1	0.793,0.0	1,1,1,1
1	EDT2	1.0	EDT2	0.756,0.0	1,1,1,1
2	EDT3	1.0	EDT3	0.723,0.0	1,1,1,1
3	EDT4	1.0	EDT4	0.708,0.0	1,1,1,1
4	EDT5	1.0	EDT5	0.7,0.0	1,1,1,1
5	EDT6	1.0	EDT6	0.706,0.0	1,1,1,1

Solve and Result

sp.ED.init()

<ED> initialized in 0.0200 seconds.

True

sp.ED.run(solver='ECOS')

<ED> solved as optimal in 0.0233 seconds, converged in 11 iterations with ECOS.

True

All decision variables are collected in the dict vars.

sp.ED.vars

OrderedDict([('pg', Var: StaticGen.pg),
             ('aBus', Var: Bus.aBus),
             ('pi', Var: Bus.pi),
             ('plf', Var: Line.plf),
             ('pru', Var: StaticGen.pru),
             ('prd', Var: StaticGen.prd),
             ('prs', Var: StaticGen.prs)])

As we can see, the generator output pg is a 2D array, and the first dimension is the generator index, and the second dimension is the time slot.

sp.ED.pg.v

array([[2.  , 2.  , 2.  , 2.  , 2.  , 2.  ],
       [2.1 , 2.1 , 2.1 , 2.1 , 2.1 , 2.1 ],
       [3.23, 2.86, 2.53, 2.38, 2.3 , 2.36],
       [0.6 , 0.6 , 0.6 , 0.6 , 0.6 , 0.6 ]])

Partial results can be accessed with desired time slot. In the retrieved result, the first dimension is the generator index, and the second dimension is the time slot.

sp.ED.get(src='pg', attr='v', idx='PV_1', horizon=['EDT1'])

2.099999999839607

Or, get multiple variables in mutliple time slots.

sp.ED.get(src='pg', attr='v', idx=['PV_1', 'PV_3'], horizon=['EDT1', 'EDT2', 'EDT3'])

array([[2.1 , 2.1 , 2.1 ],
       [3.23, 2.86, 2.53]])