branch flow model...branch flow model: relaxations, convexification, computation\爀屮may 24, 2012:...
TRANSCRIPT
![Page 1: Branch Flow Model...Branch Flow Model: relaxations, convexification, computation\爀屮May 24, 2012: Optimization and Control for Smart Grids, 32nd CN\ S Annual Conference, Santa Fe,](https://reader036.vdocuments.us/reader036/viewer/2022071507/6127c9b54d92cd56e155d16b/html5/thumbnails/1.jpg)
Branch Flow Model relaxations, convexification, computation
Masoud Farivar Steven Low
Subhonmesh Bose Lina Li Mani Chandy
Computing + Math SciencesElectrical Engineering
Caltech
June 2012
![Page 2: Branch Flow Model...Branch Flow Model: relaxations, convexification, computation\爀屮May 24, 2012: Optimization and Control for Smart Grids, 32nd CN\ S Annual Conference, Santa Fe,](https://reader036.vdocuments.us/reader036/viewer/2022071507/6127c9b54d92cd56e155d16b/html5/thumbnails/2.jpg)
Motivations
![Page 3: Branch Flow Model...Branch Flow Model: relaxations, convexification, computation\爀屮May 24, 2012: Optimization and Control for Smart Grids, 32nd CN\ S Annual Conference, Santa Fe,](https://reader036.vdocuments.us/reader036/viewer/2022071507/6127c9b54d92cd56e155d16b/html5/thumbnails/3.jpg)
DER: PVs, wind turbines, EVs, batteries, DR loads
Millions of active endpoints introducing
rapid large random fluctuations
in supply and demand
Large active network of DER
![Page 4: Branch Flow Model...Branch Flow Model: relaxations, convexification, computation\爀屮May 24, 2012: Optimization and Control for Smart Grids, 32nd CN\ S Annual Conference, Santa Fe,](https://reader036.vdocuments.us/reader036/viewer/2022071507/6127c9b54d92cd56e155d16b/html5/thumbnails/4.jpg)
ImplicationsCurrent control paradigm works well today
Low uncertainty, few active assets to control
Centralized, open-loop, human-in-loop, worst-case preventive
Schedule supplies to match loads
Future needs
Fast computation to cope with rapid, random, large fluctuations in supply, demand, voltage, freq
Simple algorithms to scale to large networks of active DER
Real-time data for adaptive control, e.g. real-time DR
![Page 5: Branch Flow Model...Branch Flow Model: relaxations, convexification, computation\爀屮May 24, 2012: Optimization and Control for Smart Grids, 32nd CN\ S Annual Conference, Santa Fe,](https://reader036.vdocuments.us/reader036/viewer/2022071507/6127c9b54d92cd56e155d16b/html5/thumbnails/5.jpg)
Our approach: endpoint controlEndpoint based control
Self-manage through local sensing, communication, control
Real-time, scalable, closed-loop, distributed, robust
Local algorithms with global perspective
Holistic framework with global objectives
Decompose global objectives into local algorithms
Control and optimization framework
Theoretical foundation for a holistic framework that integrates engineering + economics
Systematic algorithm design
GENI project:
![Page 6: Branch Flow Model...Branch Flow Model: relaxations, convexification, computation\爀屮May 24, 2012: Optimization and Control for Smart Grids, 32nd CN\ S Annual Conference, Santa Fe,](https://reader036.vdocuments.us/reader036/viewer/2022071507/6127c9b54d92cd56e155d16b/html5/thumbnails/6.jpg)
Key challengesNonconvexity
Convex relaxations
Large scale
Distributed algorithms
Uncertainty
Risk-limiting approach
![Page 7: Branch Flow Model...Branch Flow Model: relaxations, convexification, computation\爀屮May 24, 2012: Optimization and Control for Smart Grids, 32nd CN\ S Annual Conference, Santa Fe,](https://reader036.vdocuments.us/reader036/viewer/2022071507/6127c9b54d92cd56e155d16b/html5/thumbnails/7.jpg)
Why is convexity importantFoundation of LMP
Convexity justifies the use of Lagrange multipliers as various prices
Critical for efficient market theory
Efficient computation
Convexity delineates computational efficiency and intractability
A lot rides on (assumed) convexity structure•engineering, economics, regulatory
![Page 8: Branch Flow Model...Branch Flow Model: relaxations, convexification, computation\爀屮May 24, 2012: Optimization and Control for Smart Grids, 32nd CN\ S Annual Conference, Santa Fe,](https://reader036.vdocuments.us/reader036/viewer/2022071507/6127c9b54d92cd56e155d16b/html5/thumbnails/8.jpg)
Optimal power flow (OPF)Problem formulation
Carpentier 1962
Computational techniques
Dommel & Tinney 1968
Surveys: Huneault et al 1991, Momoh et al 2001, Pandya et al 2008
Bus injection model: SDP relaxation
Bai et al 2008, 2009, Lavaei et al 2010, 2012
Bose et al 2011, Zhang et al 2011, Sojoudi et al 2012
Lesieutre et al 2011
Branch flow model: SOCP relaxation
Baran & Wu 1989, Chiang & Baran 1990, Taylor 2011, Farivar et al 2011
![Page 9: Branch Flow Model...Branch Flow Model: relaxations, convexification, computation\爀屮May 24, 2012: Optimization and Control for Smart Grids, 32nd CN\ S Annual Conference, Santa Fe,](https://reader036.vdocuments.us/reader036/viewer/2022071507/6127c9b54d92cd56e155d16b/html5/thumbnails/9.jpg)
Application: Volt/VAR control
Motivation
Static capacitor control cannot cope with rapid random fluctuations of PVs on distr circuits
Inverter control
Much faster & more frequent
IEEE 1547 does not optimize VAR currently (unity PF)
![Page 10: Branch Flow Model...Branch Flow Model: relaxations, convexification, computation\爀屮May 24, 2012: Optimization and Control for Smart Grids, 32nd CN\ S Annual Conference, Santa Fe,](https://reader036.vdocuments.us/reader036/viewer/2022071507/6127c9b54d92cd56e155d16b/html5/thumbnails/10.jpg)
Load and Solar Variation
Empirical distribution of (load, solar) for Calabash
pic
pig
![Page 11: Branch Flow Model...Branch Flow Model: relaxations, convexification, computation\爀屮May 24, 2012: Optimization and Control for Smart Grids, 32nd CN\ S Annual Conference, Santa Fe,](https://reader036.vdocuments.us/reader036/viewer/2022071507/6127c9b54d92cd56e155d16b/html5/thumbnails/11.jpg)
Summary
• More reliable operation• Energy savings
![Page 12: Branch Flow Model...Branch Flow Model: relaxations, convexification, computation\爀屮May 24, 2012: Optimization and Control for Smart Grids, 32nd CN\ S Annual Conference, Santa Fe,](https://reader036.vdocuments.us/reader036/viewer/2022071507/6127c9b54d92cd56e155d16b/html5/thumbnails/12.jpg)
Theory summary
details
![Page 13: Branch Flow Model...Branch Flow Model: relaxations, convexification, computation\爀屮May 24, 2012: Optimization and Control for Smart Grids, 32nd CN\ S Annual Conference, Santa Fe,](https://reader036.vdocuments.us/reader036/viewer/2022071507/6127c9b54d92cd56e155d16b/html5/thumbnails/13.jpg)
Solution strategy
OPFnonconvex
OPF-arnonconvex
OPF-crconvex
exactrelaxation
inverseprojection
for treeangle
relaxation
SOCPrelaxation
![Page 14: Branch Flow Model...Branch Flow Model: relaxations, convexification, computation\爀屮May 24, 2012: Optimization and Control for Smart Grids, 32nd CN\ S Annual Conference, Santa Fe,](https://reader036.vdocuments.us/reader036/viewer/2022071507/6127c9b54d92cd56e155d16b/html5/thumbnails/14.jpg)
OPF solution
Solve OPF‐cr
OPF solution
Recover angles
radial
SOCP
• explicit formula• distributed alg
![Page 15: Branch Flow Model...Branch Flow Model: relaxations, convexification, computation\爀屮May 24, 2012: Optimization and Control for Smart Grids, 32nd CN\ S Annual Conference, Santa Fe,](https://reader036.vdocuments.us/reader036/viewer/2022071507/6127c9b54d92cd56e155d16b/html5/thumbnails/15.jpg)
OPF solution
Solve OPF‐cr
???
N
OPF solution
Recover angles
radial
angle recoverycondition holds? Ymesh
![Page 16: Branch Flow Model...Branch Flow Model: relaxations, convexification, computation\爀屮May 24, 2012: Optimization and Control for Smart Grids, 32nd CN\ S Annual Conference, Santa Fe,](https://reader036.vdocuments.us/reader036/viewer/2022071507/6127c9b54d92cd56e155d16b/html5/thumbnails/16.jpg)
Recap: solution strategy
OPFnonconvex
OPF-arnonconvex
OPF-crconvex
exactrelaxation
inverseprojection
for treeangle
relaxation
SOCPrelaxation
??
![Page 17: Branch Flow Model...Branch Flow Model: relaxations, convexification, computation\爀屮May 24, 2012: Optimization and Control for Smart Grids, 32nd CN\ S Annual Conference, Santa Fe,](https://reader036.vdocuments.us/reader036/viewer/2022071507/6127c9b54d92cd56e155d16b/html5/thumbnails/17.jpg)
OPF solution
Solve OPF‐cr
Optimize phaseshifters
N
OPF solution
Recover angles
radial
angle recoverycondition holds? Ymesh
design forsimplicity
![Page 18: Branch Flow Model...Branch Flow Model: relaxations, convexification, computation\爀屮May 24, 2012: Optimization and Control for Smart Grids, 32nd CN\ S Annual Conference, Santa Fe,](https://reader036.vdocuments.us/reader036/viewer/2022071507/6127c9b54d92cd56e155d16b/html5/thumbnails/18.jpg)
Examples
With PS
![Page 19: Branch Flow Model...Branch Flow Model: relaxations, convexification, computation\爀屮May 24, 2012: Optimization and Control for Smart Grids, 32nd CN\ S Annual Conference, Santa Fe,](https://reader036.vdocuments.us/reader036/viewer/2022071507/6127c9b54d92cd56e155d16b/html5/thumbnails/19.jpg)
Examples
With PS
![Page 20: Branch Flow Model...Branch Flow Model: relaxations, convexification, computation\爀屮May 24, 2012: Optimization and Control for Smart Grids, 32nd CN\ S Annual Conference, Santa Fe,](https://reader036.vdocuments.us/reader036/viewer/2022071507/6127c9b54d92cd56e155d16b/html5/thumbnails/20.jpg)
Extension: distributed solution
Work in progress with Lina Li, Lingwen Gan, Caltech
Local algorithm at bus jupdate local variables based on Lagrange multipliers from childrensend Lagrange multipliers to parents
i
local Lagrange multipliers
vj
Pij,Qij, ij
local load, generation
highly parallelizable !
![Page 21: Branch Flow Model...Branch Flow Model: relaxations, convexification, computation\爀屮May 24, 2012: Optimization and Control for Smart Grids, 32nd CN\ S Annual Conference, Santa Fe,](https://reader036.vdocuments.us/reader036/viewer/2022071507/6127c9b54d92cd56e155d16b/html5/thumbnails/21.jpg)
Extension: distributed solution
Work in progress with Lina Li, Lingwen Gan, Caltech
TheoremDistributed algorithm converges
to global optimal for radial networks
to global optimal for convexified mesh networks
to approximate/optimal for general mesh networks
4 5
00.10.20.3 p (MW)
140145150155
0123
0 1000 2000 3000 4000
3
1.52
2.5 P0
(MW)
P (MW)
SCE distribution circuit
![Page 22: Branch Flow Model...Branch Flow Model: relaxations, convexification, computation\爀屮May 24, 2012: Optimization and Control for Smart Grids, 32nd CN\ S Annual Conference, Santa Fe,](https://reader036.vdocuments.us/reader036/viewer/2022071507/6127c9b54d92cd56e155d16b/html5/thumbnails/22.jpg)
Key messages
Radial networks computationally simple
Exploit tree graph & convex relaxation
Real-time scalable control promising
Mesh networks can be convexified
Design for simplicity
Need few (?) phase shifters (sparse topology)
Efficient algorithms for mesh without phase shifters (work in progress)
![Page 23: Branch Flow Model...Branch Flow Model: relaxations, convexification, computation\爀屮May 24, 2012: Optimization and Control for Smart Grids, 32nd CN\ S Annual Conference, Santa Fe,](https://reader036.vdocuments.us/reader036/viewer/2022071507/6127c9b54d92cd56e155d16b/html5/thumbnails/23.jpg)
Theory: details
Branch flow model and OPF
Solution strategy: two relaxations
Angle relaxation
SOCP relaxation
Convexification for mesh networks
Extensions
![Page 24: Branch Flow Model...Branch Flow Model: relaxations, convexification, computation\爀屮May 24, 2012: Optimization and Control for Smart Grids, 32nd CN\ S Annual Conference, Santa Fe,](https://reader036.vdocuments.us/reader036/viewer/2022071507/6127c9b54d92cd56e155d16b/html5/thumbnails/24.jpg)
Two models
i j k
sjg s j
c
SijSjkbranch
flow
Sj Sjkkbus
injection
![Page 25: Branch Flow Model...Branch Flow Model: relaxations, convexification, computation\爀屮May 24, 2012: Optimization and Control for Smart Grids, 32nd CN\ S Annual Conference, Santa Fe,](https://reader036.vdocuments.us/reader036/viewer/2022071507/6127c9b54d92cd56e155d16b/html5/thumbnails/25.jpg)
Two models
i j kzij
Vi VjIij
branch current I j I jk
k
buscurrent
![Page 26: Branch Flow Model...Branch Flow Model: relaxations, convexification, computation\爀屮May 24, 2012: Optimization and Control for Smart Grids, 32nd CN\ S Annual Conference, Santa Fe,](https://reader036.vdocuments.us/reader036/viewer/2022071507/6127c9b54d92cd56e155d16b/html5/thumbnails/26.jpg)
Two models
Vi Vj
Si ViIi
*
Sij ViIij*
Equivalent models of Kirchhoff laws
Bus injection model focuses on nodal vars
Branch flow model focuses on branch vars
![Page 27: Branch Flow Model...Branch Flow Model: relaxations, convexification, computation\爀屮May 24, 2012: Optimization and Control for Smart Grids, 32nd CN\ S Annual Conference, Santa Fe,](https://reader036.vdocuments.us/reader036/viewer/2022071507/6127c9b54d92cd56e155d16b/html5/thumbnails/27.jpg)
Two models
Vi Vj
Si ViIi
*
Sij ViIij*
1. What is the model?2. What is OPF in the model?3. What is the solution strategy?
![Page 28: Branch Flow Model...Branch Flow Model: relaxations, convexification, computation\爀屮May 24, 2012: Optimization and Control for Smart Grids, 32nd CN\ S Annual Conference, Santa Fe,](https://reader036.vdocuments.us/reader036/viewer/2022071507/6127c9b54d92cd56e155d16b/html5/thumbnails/28.jpg)
let’s start with something familiar
![Page 29: Branch Flow Model...Branch Flow Model: relaxations, convexification, computation\爀屮May 24, 2012: Optimization and Control for Smart Grids, 32nd CN\ S Annual Conference, Santa Fe,](https://reader036.vdocuments.us/reader036/viewer/2022071507/6127c9b54d92cd56e155d16b/html5/thumbnails/29.jpg)
Bus injection model
Sj VjI j
* for all jI YVSj sj for all j
admittance matrix:
Yij :
yikk~i if i j
yij if i ~ j0 else
Sj VjI j
*
sj sjc sj
g
Kirchhoff law
power balance
power definition
![Page 30: Branch Flow Model...Branch Flow Model: relaxations, convexification, computation\爀屮May 24, 2012: Optimization and Control for Smart Grids, 32nd CN\ S Annual Conference, Santa Fe,](https://reader036.vdocuments.us/reader036/viewer/2022071507/6127c9b54d92cd56e155d16b/html5/thumbnails/30.jpg)
Bus injection model: OPF
min f j Re Sj j
over x : S, I,V, s subject to I YV
Sj sj Sj VjI j
*
s j sj s j
V k |Vk | V k
e.g. generation cost
Kirchhoff law
power balance
![Page 31: Branch Flow Model...Branch Flow Model: relaxations, convexification, computation\爀屮May 24, 2012: Optimization and Control for Smart Grids, 32nd CN\ S Annual Conference, Santa Fe,](https://reader036.vdocuments.us/reader036/viewer/2022071507/6127c9b54d92cd56e155d16b/html5/thumbnails/31.jpg)
min tr MkWkG
over W positive semidefinite matrix
s.t. Pk tr kW Pk
Qk tr kW Qk
V k2 tr JkW V k
2
W 0, rank W 1
Bus injection model: OPF
convex relaxation: SDP
• only variable is matrix W
• obj & constraintsare linear in W
• … except …
![Page 32: Branch Flow Model...Branch Flow Model: relaxations, convexification, computation\爀屮May 24, 2012: Optimization and Control for Smart Grids, 32nd CN\ S Annual Conference, Santa Fe,](https://reader036.vdocuments.us/reader036/viewer/2022071507/6127c9b54d92cd56e155d16b/html5/thumbnails/32.jpg)
Bus injection model: SDRNon-convex QCQP
Rank-constrained SDP
Relax the rank constraint and solve the SDP
Does the optimal solution satisfy the rank-constraint?
We are done! Solution may notbe meaningful
yes no
Lavaei 2010, 2012Radial: Bose 2011, Zhang 2011
Sojoudi 2011Lesiertre 2011
Bai 2008
![Page 33: Branch Flow Model...Branch Flow Model: relaxations, convexification, computation\爀屮May 24, 2012: Optimization and Control for Smart Grids, 32nd CN\ S Annual Conference, Santa Fe,](https://reader036.vdocuments.us/reader036/viewer/2022071507/6127c9b54d92cd56e155d16b/html5/thumbnails/33.jpg)
Bus injection model: summary
OPF = rank constrained SDP
Sufficient conditions for SDP to be exact
Whether a solution is globally optimal is always easily checkable
Mesh: must solve SDP to check
Tree: depends only on constraint pattern or r/x ratios
![Page 34: Branch Flow Model...Branch Flow Model: relaxations, convexification, computation\爀屮May 24, 2012: Optimization and Control for Smart Grids, 32nd CN\ S Annual Conference, Santa Fe,](https://reader036.vdocuments.us/reader036/viewer/2022071507/6127c9b54d92cd56e155d16b/html5/thumbnails/34.jpg)
Two models
Vi Vj
Si ViIi
*
Sij ViIij*
1. What is the model?2. What is OPF in the model?3. What is the solution strategy?
![Page 35: Branch Flow Model...Branch Flow Model: relaxations, convexification, computation\爀屮May 24, 2012: Optimization and Control for Smart Grids, 32nd CN\ S Annual Conference, Santa Fe,](https://reader036.vdocuments.us/reader036/viewer/2022071507/6127c9b54d92cd56e155d16b/html5/thumbnails/35.jpg)
Branch flow model
Ohm’s law
Sij ViIij* for all i j
Vi Vj zij Iij for all i j
Sij zij Iij2
i j Sjk
jk sj for all j power balance
sj
sendingend pwr loss
sendingend pwr
power def
![Page 36: Branch Flow Model...Branch Flow Model: relaxations, convexification, computation\爀屮May 24, 2012: Optimization and Control for Smart Grids, 32nd CN\ S Annual Conference, Santa Fe,](https://reader036.vdocuments.us/reader036/viewer/2022071507/6127c9b54d92cd56e155d16b/html5/thumbnails/36.jpg)
Sij ViIij*
Sij Sjkk: j~k zij Iij
2 sj
c sjgKirchoff’s Law:
Vj Vi zij IijOhm’s Law:
min riji~ j Iij
2 i
i Vi
2
over (S, I,V, sg, sc )
s. t. sig si
g sig si si
c vi vi vi
Branch flow model: OPF
real power loss CVR (conservationvoltage reduction)
![Page 37: Branch Flow Model...Branch Flow Model: relaxations, convexification, computation\爀屮May 24, 2012: Optimization and Control for Smart Grids, 32nd CN\ S Annual Conference, Santa Fe,](https://reader036.vdocuments.us/reader036/viewer/2022071507/6127c9b54d92cd56e155d16b/html5/thumbnails/37.jpg)
Sij ViIij*Vj Vi zij Iij
branch flowmodel
OPF
Sij zij Iij2
i j Sjk
jk sj
c sjg
min f h(x) over x : (S, I,V, sg, sc )
s. t. sig si
g sig si si
c vi vi vi
generation,VAR control
![Page 38: Branch Flow Model...Branch Flow Model: relaxations, convexification, computation\爀屮May 24, 2012: Optimization and Control for Smart Grids, 32nd CN\ S Annual Conference, Santa Fe,](https://reader036.vdocuments.us/reader036/viewer/2022071507/6127c9b54d92cd56e155d16b/html5/thumbnails/38.jpg)
Sij ViIij*Vj Vi zij Iij
branch flowmodel
OPF
Sij zij Iij2
i j Sjk
jk sj
c sjg
min f h(x) over x : (S, I,V, sg, sc )
s. t. sig si
g sig si si
c vi vi vi
demandresponse
![Page 39: Branch Flow Model...Branch Flow Model: relaxations, convexification, computation\爀屮May 24, 2012: Optimization and Control for Smart Grids, 32nd CN\ S Annual Conference, Santa Fe,](https://reader036.vdocuments.us/reader036/viewer/2022071507/6127c9b54d92cd56e155d16b/html5/thumbnails/39.jpg)
Theory
Branch flow model and OPF
Solution strategy: two relaxations
Angle relaxation
SOCP relaxation
Convexification for mesh networks
Extensions
![Page 40: Branch Flow Model...Branch Flow Model: relaxations, convexification, computation\爀屮May 24, 2012: Optimization and Control for Smart Grids, 32nd CN\ S Annual Conference, Santa Fe,](https://reader036.vdocuments.us/reader036/viewer/2022071507/6127c9b54d92cd56e155d16b/html5/thumbnails/40.jpg)
Solution strategy
OPFnonconvex
OPF-arnonconvex
OPF-crconvex
exactrelaxation
inverseprojection
for treeangle
relaxation
SOCPrelaxation
![Page 41: Branch Flow Model...Branch Flow Model: relaxations, convexification, computation\爀屮May 24, 2012: Optimization and Control for Smart Grids, 32nd CN\ S Annual Conference, Santa Fe,](https://reader036.vdocuments.us/reader036/viewer/2022071507/6127c9b54d92cd56e155d16b/html5/thumbnails/41.jpg)
Sij ViIij*Vj Vi zij Iij
branch flowmodel
OPF
Sij zij Iij2
i j Sjk
jk sj
c sjg
min f h(x) over x : (S, I,V, sg, sc )
s. t. sig si
g sig si si
c vi vi vi
![Page 42: Branch Flow Model...Branch Flow Model: relaxations, convexification, computation\爀屮May 24, 2012: Optimization and Control for Smart Grids, 32nd CN\ S Annual Conference, Santa Fe,](https://reader036.vdocuments.us/reader036/viewer/2022071507/6127c9b54d92cd56e155d16b/html5/thumbnails/42.jpg)
OPF
x X X
min f h(x) over x : (S, I,V, sg, sc )
s. t. sig si
g sig si si
c vi vi vi
![Page 43: Branch Flow Model...Branch Flow Model: relaxations, convexification, computation\爀屮May 24, 2012: Optimization and Control for Smart Grids, 32nd CN\ S Annual Conference, Santa Fe,](https://reader036.vdocuments.us/reader036/viewer/2022071507/6127c9b54d92cd56e155d16b/html5/thumbnails/43.jpg)
min f y over y : (S,,v, sg, sc )
s. t. sig si
g sig si si
c vi vi vi
OPF-ar
y : h(x) Y
Y
Xrelax each voltage/current from a
point in complex plane into a circle
![Page 44: Branch Flow Model...Branch Flow Model: relaxations, convexification, computation\爀屮May 24, 2012: Optimization and Control for Smart Grids, 32nd CN\ S Annual Conference, Santa Fe,](https://reader036.vdocuments.us/reader036/viewer/2022071507/6127c9b54d92cd56e155d16b/html5/thumbnails/44.jpg)
min f y over y : (S,,v, sg, sc )
s. t. sig si
g sig si si
c vi vi vi
OPF-cr
y conv Y
Y
X
relax to convex hull(SOCP)
![Page 45: Branch Flow Model...Branch Flow Model: relaxations, convexification, computation\爀屮May 24, 2012: Optimization and Control for Smart Grids, 32nd CN\ S Annual Conference, Santa Fe,](https://reader036.vdocuments.us/reader036/viewer/2022071507/6127c9b54d92cd56e155d16b/html5/thumbnails/45.jpg)
Recap so far …
OPFnonconvex
OPF-arnonconvex
OPF-crconvex
exactrelaxation
inverseprojection
for treeangle
relaxation
SOCPrelaxation
![Page 46: Branch Flow Model...Branch Flow Model: relaxations, convexification, computation\爀屮May 24, 2012: Optimization and Control for Smart Grids, 32nd CN\ S Annual Conference, Santa Fe,](https://reader036.vdocuments.us/reader036/viewer/2022071507/6127c9b54d92cd56e155d16b/html5/thumbnails/46.jpg)
TheoremOPF-cr is convex (SOCP)
much simpler than SDP
OPF-cr is exact relaxation
OPF-cr is exact
optimal of OPF-cr is also optimal for OPF-ar
for mesh as well as radial networks
real & reactive powers, but volt/current mags
![Page 47: Branch Flow Model...Branch Flow Model: relaxations, convexification, computation\爀屮May 24, 2012: Optimization and Control for Smart Grids, 32nd CN\ S Annual Conference, Santa Fe,](https://reader036.vdocuments.us/reader036/viewer/2022071507/6127c9b54d92cd56e155d16b/html5/thumbnails/47.jpg)
OPF ??
Angle recovery
Y
y
OPF-arh1(y) X ?
Y
X
Y
X
y y
does there exist s.t.
![Page 48: Branch Flow Model...Branch Flow Model: relaxations, convexification, computation\爀屮May 24, 2012: Optimization and Control for Smart Grids, 32nd CN\ S Annual Conference, Santa Fe,](https://reader036.vdocuments.us/reader036/viewer/2022071507/6127c9b54d92cd56e155d16b/html5/thumbnails/48.jpg)
TheoremInverse projection exist iff s.t.
Angle recovery
B y
Two simple angle recovery algorithms
centralized: explicit formula
decentralized: recursive alg
!
incidence matrix;depends on topology
depends on OPF-ar solution
![Page 49: Branch Flow Model...Branch Flow Model: relaxations, convexification, computation\爀屮May 24, 2012: Optimization and Control for Smart Grids, 32nd CN\ S Annual Conference, Santa Fe,](https://reader036.vdocuments.us/reader036/viewer/2022071507/6127c9b54d92cd56e155d16b/html5/thumbnails/49.jpg)
TheoremFor radial network:
Angle recovery
B y !
X Yy
Y
X
y
mesh tree
![Page 50: Branch Flow Model...Branch Flow Model: relaxations, convexification, computation\爀屮May 24, 2012: Optimization and Control for Smart Grids, 32nd CN\ S Annual Conference, Santa Fe,](https://reader036.vdocuments.us/reader036/viewer/2022071507/6127c9b54d92cd56e155d16b/html5/thumbnails/50.jpg)
OPF solution
Solve OPF‐cr
OPF solution
Recover angles
radial
SOCP
• explicit formula• distributed alg
![Page 51: Branch Flow Model...Branch Flow Model: relaxations, convexification, computation\爀屮May 24, 2012: Optimization and Control for Smart Grids, 32nd CN\ S Annual Conference, Santa Fe,](https://reader036.vdocuments.us/reader036/viewer/2022071507/6127c9b54d92cd56e155d16b/html5/thumbnails/51.jpg)
OPF solution
Solve OPF‐cr
???
N
OPF solution
Recover angles
radial
angle recoverycondition holds? Ymesh
![Page 52: Branch Flow Model...Branch Flow Model: relaxations, convexification, computation\爀屮May 24, 2012: Optimization and Control for Smart Grids, 32nd CN\ S Annual Conference, Santa Fe,](https://reader036.vdocuments.us/reader036/viewer/2022071507/6127c9b54d92cd56e155d16b/html5/thumbnails/52.jpg)
Outline
Branch flow model and OPF
Solution strategy: two relaxations
Angle relaxation
SOCP relaxation
Convexification for mesh networks
Extensions
![Page 53: Branch Flow Model...Branch Flow Model: relaxations, convexification, computation\爀屮May 24, 2012: Optimization and Control for Smart Grids, 32nd CN\ S Annual Conference, Santa Fe,](https://reader036.vdocuments.us/reader036/viewer/2022071507/6127c9b54d92cd56e155d16b/html5/thumbnails/53.jpg)
Recap: solution strategy
OPFnonconvex
OPF-arnonconvex
OPF-crconvex
exactrelaxation
inverseprojection
for treeangle
relaxation
SOCPrelaxation
??
![Page 54: Branch Flow Model...Branch Flow Model: relaxations, convexification, computation\爀屮May 24, 2012: Optimization and Control for Smart Grids, 32nd CN\ S Annual Conference, Santa Fe,](https://reader036.vdocuments.us/reader036/viewer/2022071507/6127c9b54d92cd56e155d16b/html5/thumbnails/54.jpg)
Convexification of mesh networks
OPF minx
f h(x) s.t. x X
Theorem••Need phase shifters only
outside spanning tree
X Y
OPF-ar minx
f h(x) s.t. x Y
Y
X
OPF-ps minx,
f h(x) s.t. x, X
X
X
optimize over phase shifters as well
![Page 55: Branch Flow Model...Branch Flow Model: relaxations, convexification, computation\爀屮May 24, 2012: Optimization and Control for Smart Grids, 32nd CN\ S Annual Conference, Santa Fe,](https://reader036.vdocuments.us/reader036/viewer/2022071507/6127c9b54d92cd56e155d16b/html5/thumbnails/55.jpg)
OPF solution
Solve OPF‐cr
Optimize phaseshifters
N
OPF solution
Recover angles
radial
angle recoverycondition holds? Ymesh
• explicit formula• distributed alg
![Page 56: Branch Flow Model...Branch Flow Model: relaxations, convexification, computation\爀屮May 24, 2012: Optimization and Control for Smart Grids, 32nd CN\ S Annual Conference, Santa Fe,](https://reader036.vdocuments.us/reader036/viewer/2022071507/6127c9b54d92cd56e155d16b/html5/thumbnails/56.jpg)
Outline
Branch flow model and OPF
Solution strategy: two relaxations
Angle relaxation
SOCP relaxation
Convexification for mesh networks
Extensions
![Page 57: Branch Flow Model...Branch Flow Model: relaxations, convexification, computation\爀屮May 24, 2012: Optimization and Control for Smart Grids, 32nd CN\ S Annual Conference, Santa Fe,](https://reader036.vdocuments.us/reader036/viewer/2022071507/6127c9b54d92cd56e155d16b/html5/thumbnails/57.jpg)
Extension: equivalence
Work in progress with Subhonmesh Bose, Mani Chandy
TheoremBI and BF model are equivalent(there is a bijection between and )X X
X : x S, I,V BI model X : x S, I,V BF model
![Page 58: Branch Flow Model...Branch Flow Model: relaxations, convexification, computation\爀屮May 24, 2012: Optimization and Control for Smart Grids, 32nd CN\ S Annual Conference, Santa Fe,](https://reader036.vdocuments.us/reader036/viewer/2022071507/6127c9b54d92cd56e155d16b/html5/thumbnails/58.jpg)
Extension: equivalence
Work in progress with Subhonmesh Bose, Mani Chandy
Theorem: radial networks in SOCP W in SDR satisfies angle cond W has rank 1
y
SDR W 0 SOCP y : S,, v
y g W
W g1 y
y
![Page 59: Branch Flow Model...Branch Flow Model: relaxations, convexification, computation\爀屮May 24, 2012: Optimization and Control for Smart Grids, 32nd CN\ S Annual Conference, Santa Fe,](https://reader036.vdocuments.us/reader036/viewer/2022071507/6127c9b54d92cd56e155d16b/html5/thumbnails/59.jpg)
Backup Slides
![Page 60: Branch Flow Model...Branch Flow Model: relaxations, convexification, computation\爀屮May 24, 2012: Optimization and Control for Smart Grids, 32nd CN\ S Annual Conference, Santa Fe,](https://reader036.vdocuments.us/reader036/viewer/2022071507/6127c9b54d92cd56e155d16b/html5/thumbnails/60.jpg)
PCPM algorithmProblem:
Idea: Introduce Lagaranian variables y for Ax=zSolve the saddle point of the Lagaranian function L(x,y,z)
Algorithm :
For each step k,
Predictor :
Correcto r:
Predictor Corrector Proximal Multiplier (PCPM) method
![Page 61: Branch Flow Model...Branch Flow Model: relaxations, convexification, computation\爀屮May 24, 2012: Optimization and Control for Smart Grids, 32nd CN\ S Annual Conference, Santa Fe,](https://reader036.vdocuments.us/reader036/viewer/2022071507/6127c9b54d92cd56e155d16b/html5/thumbnails/61.jpg)
using branch flow in tree networks
Linear coupled local constraints
Ax=z
Local constraints
Local variables for each bus i: Corresponding local Lagrangian multipliers
![Page 62: Branch Flow Model...Branch Flow Model: relaxations, convexification, computation\爀屮May 24, 2012: Optimization and Control for Smart Grids, 32nd CN\ S Annual Conference, Santa Fe,](https://reader036.vdocuments.us/reader036/viewer/2022071507/6127c9b54d92cd56e155d16b/html5/thumbnails/62.jpg)
Case study
45
Schematic Diagram of a SCE distribution System
![Page 63: Branch Flow Model...Branch Flow Model: relaxations, convexification, computation\爀屮May 24, 2012: Optimization and Control for Smart Grids, 32nd CN\ S Annual Conference, Santa Fe,](https://reader036.vdocuments.us/reader036/viewer/2022071507/6127c9b54d92cd56e155d16b/html5/thumbnails/63.jpg)
0
0.1
0.2
0.3 p (MW)
140
145
150
155
0
1
2
3
0 1000 2000 3000 4000
3
1.5
2
2.5P0 (MW)
P (MW)