Download - AMul&’ScaleDesignandControlFrameworkfor ...stein/Research-NSF-V2G/efri... · AMul&’Scale"Design"and"Control"Frameworkfor"" DynamicallyCoupledSustainableandResilientInfrastructures,"

Transcript
Page 1: AMul&’ScaleDesignandControlFrameworkfor ...stein/Research-NSF-V2G/efri... · AMul&’Scale"Design"and"Control"Frameworkfor"" DynamicallyCoupledSustainableandResilientInfrastructures,"

A  Mul&-­‐Scale  Design  and  Control  Framework  for    Dynamically  Coupled  Sustainable  and  Resilient  Infrastructures,  

 with  Applica&on  to  Vehicle-­‐to-­‐Grid  Integra&on  (EFRI  –  RESIN,  Award  Number:  0835995)  

People  PI:  Jeffrey  Stein  [[email protected]]  (University  of  Michigan)  Co-­‐PIs:  Zoran  Filipi  [[email protected]],  Gregory  Keoleian  [[email protected]],  Huei  Peng  [[email protected]]  (U-­‐M)  and  Mariesa  Crow  [[email protected]]  (Missouri  University  of  Science  and  Technology)    Par&cipa&ng  Inves&gators:  Duncan  Callaway  [[email protected]]  (Berkeley),    Hosam  Fathy  [[email protected]]  (U-­‐M),  Carl  Simon  [[email protected]]  (U-­‐M),  John  Sullivan  [[email protected]]  (U-­‐M)  and  Jing  Sun  [[email protected]]  (U-­‐M)  

Good  and  frequent  communicaRon  and  interacRon  is  considered  key  to  our   project's   success   and   this   occurs   primarily   at   biweekly   research  meeRngs.    At  each  meeRng  a  presentaRon  is  given  by  of  one  of  the  task  area   researchers.     This   allows   team  members   see   what   each   task   is  focused   on,  what   their   results  mean,   how   it   affects   all   tasks   projects,  and  the  potenRal  for  task  integraRon.    Offsite  team  members  aVend  via  teleconferencing   and   receive   slides   via   email.     Offsite   team  members  can  visit  U  of  M  as  oWen  and  for  as  long  as  they  would  to  because  office  space  has  been  made  available  for  them.  

Research  Objec&ves   Project  Descrip&on  

Enabling  Poten&ally  Transforma&ve  Results  Status  of  Research  

Managing  a  Mul&-­‐Disciplinary  and  Mul&-­‐Ins&tu&onal  Project  

Vehicle  to  Grid  (V2G)  may  poten&ally  improve  resilience  by:      -­‐CreaRng  a  redundancy  of  power  sources  and  flow  paths.      -­‐Improving  grid  integrity  to  disturbances  through  energy  storage.      -­‐Decreasing  the  load  through  peak  shaving  and  reacRve  power.  

V2G  provides  sustainability  through  increased  energy  storage:      -­‐Allowing  the  grid  to  beVer  absorb  renewable  electricity.      -­‐RedistribuRng  power  demand  over  Rme  in  both  infrastructures.      -­‐Possibly  decreasing  use  of  expensive  grid  peaking  units.  

Sustainability  and  Resilience  

FUEL  PUMP  

PHEV  

BATTERY  Outlet  

Personal  TransportaRon  Infrastructure  

Electric  Power  Infrastructure  

Mul&-­‐Role  Intermediaries  (MRIs)  An   infrastructure’s   sustainability   and  resilience  oWen  depend  on  how  strongly  coupled   it   is   to   other   infrastructures  through   the   exchange   of   commodiRes,  resources,  services,  or  informaRon.  This  exchange   oWen   takes   place   through  MulR-­‐Role  Intermediaries  which  may  be  organizaRons,   individuals   or   intelligent  devices.  

Plug-­‐in   hybrid   electric   vehicles   are   an  important  MRI  because  they  couple  the  personal   transportaRon   infrastructure  with   the   electric   power   infrastructure.    This  is  the  project’s  test  bed  applicaRon.  

0  2000  4000  6000  8000  

10000  12000  14000  16000  

0   1   2   3   4   5   6   7   8   9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  

deman

d,  M

W  

hour  of  day  

Electricity  Demand,  Effect  of  DSM  

non-­‐PHEV  demand  

total  demand,  no  DSM  

total  demand,  w/  DSM  50%  

total  demand  w/  DSM  100%  

No  DSM  

50%  DSM  

100%  DSM  

PHEV  charging,  gCO2/mi   43   41   41  

Tailpipe  GHGs,  gCO2/mi    (Fleet  avg)   258   254   254  

5.9  Michigan  million  drivers,  50%  of  vehicles  are  PHEVs  

Monte  Carlo  SimulaRon  for  Lyapunov  Stability  

Angle  

Lyapunov  Energy  

AsymptoRc  bounds  for  the  probability  of  load  curtailment  for  a  large  populaRon  system  with  controllable  loads  

Time  to  ramp  MRI  response  up  and  down,  versus  probabilisRc  bound  

Probabilis&c  Control  of  Charging  PHEV  Fleets  

Demand  Side  Management  (DSM)  -­‐  X%  PHEV's  disallowed  from  charging  between  2  and  9PM  

Task  2:  Infrastructure  Resilience  Modeling  

Task  3:  MRI  Design  OpRmizaRon  

Task  1:  Infrastructure  Sustainability  Modeling  (ABM/LCA)  

Task  5:  Infrastructure  Control   Task  6:  Model  IntegraRon/ReducRon  

Above  Figure  from:  T.  Yoshida,  M.  Takahashi,  S.  Morikawa,  et  al.  2006  

Li-­‐ion  Ba\ery  Electrochemical  Health  SimulaRon  

Controlled   PHEV   charging  will   beVer   uRlize   the  generaRon   assets   and  renewable   resources   during  light   load   hours,   and   help  prevent   increases   in   peak  load  and  grid  instabiliRes.        

Total  energy  [GJ] Losses  [GJ] Peak  load  [pu] Peak  hr

Uncontrolled 19.9898 0.1284  (0.64%) 0.3350 6:12  PM

Min.  losses 19.8818 0.0373  (0.19%) 0.2457 3:04  AM

Dual  Tariff 19.9898 0.1284  (0.64%) 0.5235 8:24  PM

Task  4:  Intermediary  Control