Department of Chemical and Biological Engineering Illinois Institute of Technology On the Use of Multistage Stochastic Programming for the Design of Smart Grid Coordinated Systems Donald J. Chmielewski Oluwasanmi Adeodu and Jin Zhang Department of Chemical and Biological Engineering Illinois Institute of Technology Minimally Backed-off Operating Point Different Controller Tuning Values Expected Dynamic Operating Regions Steady-State Operating Line Optimal Steady-State Operating Point 1
43
Embed
On the Use of Multistage Stochastic Programming for the ...mypages.iit.edu/~chmielewski/presentations/2015/... · On the Use of Multistage Stochastic Programming for the Design of
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Department of Chemical and Biological Engineering
Illinois Institute of Technology
On the Use of Multistage Stochastic
Programming for the Design of Smart Grid
Coordinated Systems
Donald J. ChmielewskiOluwasanmi Adeodu and Jin Zhang
Department of Chemical and Biological EngineeringIllinois Institute of Technology
Minimally
Backed-off
Operating
Point
Different Controller
Tuning Values
Expected
Dynamic
Operating
Regions
Steady-State
Operating
Line
Optimal Steady-State
Operating Point
1
Department of Chemical and Biological Engineering
Illinois Institute of Technology
Motivation
Dispatch Capable
Generation Power Grid
Smart Grid Electric Power Network:
Demand
(Consumers)
Renewable
Generation
Responsive
Demand Energy Storage
Existing
Components
Expected
Future
Components
0 5 10 15 20
0
200
400
600
800
time (days)
Po
wer
Req
uir
ed f
rom
Dis
pa
tch
ab
le G
ener
ato
rs
(MW
)
Baseline
Baseline with Renewable Power
0 5 10 15 20
0
200
400
600
800
time (days)
Baseline with Renewable Power
Impact of Storage and DR
2
Department of Chemical and Biological Engineering
Illinois Institute of Technology
Building HVAC Systems
Analysis requires details
of operating policy
Multistage Stochastic
Programming (MSP)
framework
59 60 61 62-100
-50
0
50
100
150
200
250
Time (days)
kW
hr
/ d
ay
Heat from Room
Heat to Cooler
Heat to TES Unit
59 60 61 62-50
0
50
100
150
200
250
Time (days)
kW
hr
/ d
ay
Heat from Room
Heat to Cooler
Heat to TES Unit
3
Heat from
BuildingBuilding
Heat from
Environment
Power
Consumption Chiller
Heat to
TES
Thermal
Energy Storage
Heat to
Chiller
Department of Chemical and Biological Engineering
Illinois Institute of Technology
Integrated Gasification Combined Cycle
4
Department of Chemical and Biological Engineering
Illinois Institute of Technology
Dispatch Capable IGCC
- Respond to Market Prices - Increase Average Revenue
Electricity Price
Opportunity:
100 101 102 103 104 105 106 107 108 109
70
80
90
100
110
Time (days)
Ele
ctr
icit
y V
alu
e (
$/M
W h
r)
5
Department of Chemical and Biological Engineering
Illinois Institute of Technology
Simplified view of IGCC
Gasification Block(Includes ASU Distillation,
Gasifier and Acid Gas Removal)
Power Block(Includes Expansion Turbine,
Combustion Turbine, HRSG,
and Steam Turbine)
nASUns,A
nC
ncoal
ns,H2nH2
nG
H2 Storage(MH2)
Compressed
Air Storage(MA)
MACASU Main
Air Compressor
PGPC
PN
6
Department of Chemical and Biological Engineering
Illinois Institute of Technology
Dispatch of IGCC Power Generation
100 101 102 103 104 105 106 107 108 109
0
50
100
150
Va
lue (
$/M
W h
r)
100 101 102 103 104 105 106 107 108 109
0
500
1000
1500
Po
wer (
MW
)
100 101 102 103 104 105 106 107 108 109
0
500
1000
Time (days)
Ma
ss (
ton
nes)
Instantaneous
Average
Maximum
PG
Ce
MH2
7
Department of Chemical and Biological Engineering
Illinois Institute of Technology
Instantaneous and Average Revenue
100 101 102 103 104 105 106 107 108 109
0
50
100
150
Time (days)
Rev
en
ue (
$1
00
0/h
r)
Dispatch
No Dispatch
8
Department of Chemical and Biological Engineering
Illinois Institute of Technology
Dispatch Requires Equipment Upgrade
Gasification Block(Includes ASU Distillation,
Gasifier and Acid Gas Removal)
Power Block(Includes Expansion Turbine,
Combustion Turbine, HRSG,
and Steam Turbine)
nASUns,A
nC
ncoal
ns,H2nH2
nG
H2 Storage(MH2)
Compressed
Air Storage(MA)
MACASU Main
Air Compressor
PGPC
PN
Net Present Value analysis is required
9
Department of Chemical and Biological Engineering
Illinois Institute of Technology
Presentation Outline
Motivation for Multistage Stochastic
Programming (MSP)
Review of MSP
Proposed Solution Method for MSP
Future Directions
10
Department of Chemical and Biological Engineering
Illinois Institute of Technology
Review of Stochastic Programming
Two-stage Stochastic Program:
bAxxQxcT
x s.t )(min
)(s.t )(min)( where hWyTxyqExQ T
y
x are here-and-now (equipment) variables
y are wait-and-see (operating) variables
are random (stochastic) variables
c and q() are capital and operating costs
h() is the disturbance
ym , m = 1 … M
m , m = 1 … M
11
Department of Chemical and Biological Engineering
Illinois Institute of Technology
Review of Stochastic Programming
Scenario Based Approximation:
MmhWyTx
bAxyqpxc
mm
M
m
m
T
mm
T
yx m ...1s.t min
1,
s.t min1
M
m
kmmm
T
mmy
k
T xThWyyqpxcm
Finite support of scenarios: m , m = 1 … M
Each with outcomes: qm = q(m) and hm = h(m)
Each with a probability: pm = p(m)
Corresponding wait-and-see variables: ym , m = 1 … M
Decomposition Methods Iterate over:
12
Department of Chemical and Biological Engineering
Illinois Institute of Technology
Multistage Stochastic Programming Variables indexed in time: