slide 1 the renewables challenge: keeping the lights on while managing variability and uncertainty...
TRANSCRIPT
![Page 1: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/1.jpg)
Slide 1
The Renewables Challenge:Keeping the lights on while managing
variability and uncertainty
Princeton University Academic Mini-Reunion
October 2, 2015
Warren B. PowellPENSA Laboratory
Dept. of Operations Research andFinancial Engineering
http://energysystems.princeton.edu
![Page 2: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/2.jpg)
Outline
The renewables challenge Combining wind and solar The uncertainties of energy Three energy problems
» An energy storage problem» An energy portfolio policy model» Simulating the PJM grid using SMART-ISO
Concluding remarks
![Page 3: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/3.jpg)
Outline
The renewables challenge Combining wind and solar The uncertainties of energy Three energy problems
» An energy storage problem» An energy portfolio policy model» Simulating the PJM grid using SMART-ISO
Concluding remarks
![Page 4: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/4.jpg)
![Page 5: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/5.jpg)
![Page 6: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/6.jpg)
Wind in the U.S.
![Page 7: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/7.jpg)
![Page 8: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/8.jpg)
99.9 percent from renewables!
Fossil Backup
BatteryStorage
Wind &Solar
20 GW
750 GWhr battery!
![Page 9: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/9.jpg)
Outline
The renewables challenge Combining wind and solar The uncertainties of energy Modeling sequential decision problems under
uncertainty Three energy problems
» An energy storage problem» An energy portfolio policy model» Simulating the PJM grid using SMART-ISO
Concluding remarks
![Page 10: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/10.jpg)
Wind energy in PJM
Total PJM load plus actual wind (July)
53 wind farms
![Page 11: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/11.jpg)
Wind energy in PJM
Total PJM load plus actual wind (July)
100GW
101,000 MWhr battery$50 billion!!
Wind ~ 37 percent of total load
![Page 12: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/12.jpg)
Solar energy
Solar from all PSE&G solar farms
![Page 13: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/13.jpg)
Solar energy
Total PJM load plus factored solar (July)
Solar ~ 15 percent of total load
![Page 14: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/14.jpg)
Combining wind and solar
Mixture of wind and solar to meet July load
815,000 MWhr battery $989 billion!!
![Page 15: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/15.jpg)
Combining wind and solar
Mixture of wind and solar to meet July load
100GW
260,000 MWhr battery $130 billion!!
![Page 16: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/16.jpg)
Capacity factor analysis
Computing the “capacity factor”
Capacity
Actual
Generated windCapacity factor = 39%
Maximum capacity
![Page 17: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/17.jpg)
Capacity factor analysis
Classical analysis of renewables» Multiply installed capacity by the capacity factor
• 1 MW solar panel• Capacity factor of .25• Translates to 0.25 MW of generation
» Now, treat the .25 MW as if it is a form of conventional generation.
» This makes it possible to scale up renewables without regard to the challenges of variability and uncertainty.
» Let’s try to do better.
![Page 18: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/18.jpg)
Outline
The renewables challenge Combining wind and solar The uncertainties of energy Three energy problems
» An energy storage problem» An energy portfolio policy model» Simulating the PJM grid using SMART-ISO
Concluding remarks
![Page 19: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/19.jpg)
Northeast Reliability Councils and Interconnects
![Page 20: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/20.jpg)
![Page 21: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/21.jpg)
![Page 22: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/22.jpg)
Energy from wind
1 year
Wind power from all PJM wind farms
Jan Feb March April May June July Aug Sept Oct Nov Dec
![Page 23: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/23.jpg)
Energy from wind
30 days
Wind from all PJM wind farms
![Page 24: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/24.jpg)
Modeling wind
Forecast vs. actual for a single wind farm
Actual
Forecasted
![Page 25: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/25.jpg)
Solar energy
Princeton solar array
![Page 26: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/26.jpg)
Solar energy
Princeton solar array
![Page 27: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/27.jpg)
PSE&G solar farms
Solar output over entire year (all farms)
Sept Oct Nov Dec Jan Feb March April May June July Aug
![Page 28: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/28.jpg)
Solar from PSE&G solar farms
Solar from a single solar farm
![Page 29: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/29.jpg)
Solar from PSE&G solar farms
Within-day sample trajectories
![Page 30: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/30.jpg)
Brazil
![Page 31: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/31.jpg)
Rainfall
Foz do Iguaçu (Brazil) – 2011 through 2013
2011 2012 2013
![Page 32: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/32.jpg)
Commodity prices
The price of natural gas» Reflects global and local economies, competing global
commodities (primarily oil), policies (e.g. toward CO2), and technology (e.g. fracking).
![Page 33: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/33.jpg)
$4 /mmBTU
$120 /mmBTU!
![Page 34: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/34.jpg)
![Page 35: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/35.jpg)
Locational marginal prices on the gridLMPs – Locational marginal prices
$58.47/MW
![Page 36: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/36.jpg)
Locational marginal prices on the gridLMPs – Locational marginal prices
$977/MW !!!
![Page 37: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/37.jpg)
Locational marginal prices on the gridLMPs – Locational marginal prices
$328/MW !
![Page 38: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/38.jpg)
Locational marginal prices on the grid
$52/MWhr
![Page 39: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/39.jpg)
Uncertainty
It is important to separate:» Predictable variability
• Diurnal cycles• Large weather patterns• Major human events (Super
bowl)
» Stochastic uncertainty • Temperature deviations from
forecast• Late/early arrival of a storm• Generator failures• Wind shifts
PJM load
Aggregate solar
![Page 40: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/40.jpg)
Dealing with uncertainty
Available at energysystems.princeton.edu
![Page 41: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/41.jpg)
Outline
The renewables challenge Combining wind and solar The uncertainties of energy Three energy problems
» An energy storage problem» An energy portfolio policy model» Simulating the PJM grid using SMART-ISO
Concluding remarks
![Page 42: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/42.jpg)
Policy function approximations
Battery arbitrage – When to charge, when to discharge, given volatile LMPs
![Page 43: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/43.jpg)
1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 700.00
20.00
40.00
60.00
80.00
100.00
120.00
140.00
Grid operators require that batteries bid charge and discharge prices, an hour in advance.
We have to search for the best values for the policy parameters
DischargeCharge
Charge Discharge and .
Policy function approximations
![Page 44: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/44.jpg)
Policy function approximations
Our policy function might be the parametric model (this is nonlinear in the parameters):
charge
charge discharge
charge
1 if
( | ) 0 if
1 if
t
t t
t
p
X S p
p
Energy in storage:
Price of electricity:
![Page 45: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/45.jpg)
Policy function approximations
Finding the best policy» We need to maximize
» We cannot compute the expectation, so we run simulations:
DischargeCharge
0
max ( ) , ( | )T
tt t t
t
F C S X S
E
![Page 46: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/46.jpg)
Outline
The renewables challenge Combining wind and solar The uncertainties of energy Three energy problems
» An energy storage problem» An energy portfolio policy model» Simulating the PJM grid using SMART-ISO
Concluding remarks
![Page 47: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/47.jpg)
99.9 percent from renewables!
» What answer do we get if we model this problem more carefully?
![Page 48: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/48.jpg)
SMART-Invest
Features:» Finds the optimal mix of wind, solar and storage, in the
presence of two types of fossil generation:• Slow (steam) generation, which is planned 24 hours in
advance• Fast (turbine) generation, which is planned 1 hour in advance• Real-time ramping of all fossils within ramping limits
» Simulates entire year in hourly increments, to capture all forms of variability (except subhourly)
» Minimizes investment and operating costs, possibly including SRECs and carbon tax.
» Able to directly specify the cost of fossil generation (anticipating dramatic reduction in fossils).
» Properly accounts for the marginal cost of each unit of investment.
![Page 49: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/49.jpg)
8760
,1
min ( ) ( , ( | ))Invi
inv inv opr opr invt t t tx i I
t
C x C S X S x
SMART-Invest
The investment problem:
Investment cost in wind, solarand storage.
Capital investment cost inwind, solar and storage
( )oprtX S is the operating policy.
Operating costs of fossil generators,energy losses from storage, misc. operating costs of renewables.
Wind
Solar
![Page 50: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/50.jpg)
SMART-Invest
Stage 1:Investment
Hour 1 Hour 2 Hour 3 Hr 8758 Hr 8759 Hr 8760…
Wind CapacitySolar Capacity
Battery CapacityFossil Capacity
Wind
Solar
Find search direction
Update investments
![Page 51: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/51.jpg)
SMART-Invest
Operational planning
» Meet demand while minimizing operating costs» Observe day-ahead notification requirements for
generators» Includes reserve constraints to manage uncertainty» Meet aggregate ramping constraints (but does not
schedule individual generators)
24 hour notification of steam
1 hour notification of gas
Real-time storage and ramping decisions (in hourly increments)
36 hour planning horizon using forecasts of wind and solar
![Page 52: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/52.jpg)
SMART-Invest
Operational planning model
» Model plans using rolling 36 hour horizon» Steam plants are locked in 24 hours in advance» Gas turbines are decided 1 hour in advance
Slow fossil running units1 24 36
1 24 36Slow fossil running units
1 24 36Slow fossil running units
1 24 36Slow fossil running units
1 24 36Slow fossil running units
1 24 36Slow fossil running units
1 24 36Slow fossil running units
1 24 36Slow fossil running units
…
…
…1 2 3 4 5 8760
1 year
Lock in steam generation decisions 24 hours in advanceThe tentative plan is discarded
![Page 53: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/53.jpg)
SMART-Invest
Lookahead model with adjustment» Objective function
» Reserve constraint:
» Other constraints:• Ramping• Capacity constraints• Conservation of flow in storage• ….
Tunable policy parameter
![Page 54: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/54.jpg)
Robust policies
Policy search – Optimizing reserve parameter
Low carbon tax, increased usage of slow fossil, requires higher reserve margin ~19 percent
High carbon tax, shift from slow to fast fossil, requires minimal reserve margin ~1 pct
![Page 55: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/55.jpg)
The value of storage
The marginal value of storage» On the margin, value of storage can be expensive!
Ene
rgy
in s
tora
ge
Time
This investment in batteries is only used a small fraction of the time.
![Page 56: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/56.jpg)
Policy studies
Renewables as a function of cost of fossil fuels
Solar
Wind
$/MWhr cost of fossil fuels
100
80
60
40
20
0Perc
ent f
rom
ren
ewab
le Total renewables
0 20 50 60 70 80 90 100 150 300 400 500 1000 3000
» Study assumes unconstrained access to wind at lowest cost (this is not available in the eastern U.S.)
» Note the reluctance to introduce solar…
» … and even greater reluctance to use storage.
Battery
![Page 57: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/57.jpg)
Policy studies
Sensitivity to CO2 tax.
Slow fossil Fast fossil
Nuclear
Wind
Solar
“Other” fast fossil
![Page 58: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/58.jpg)
Outline
The renewables challenge Combining wind and solar The uncertainties of energy Three energy problems
» An energy storage problem» An energy portfolio policy model» Simulating the PJM grid using SMART-ISO
Concluding remarks
![Page 59: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/59.jpg)
© 2010 Warren B. Powell Slide 59
Lecture outline
Simulating the PJM grid with SMART-ISO
The PJM planning process Model validation Modeling wind Integrating offshore wind
![Page 60: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/60.jpg)
![Page 61: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/61.jpg)
The timing of decisions
Day-ahead planning (slow – predominantly steam)
Intermediate-term planning (fast – gas turbines)
Real-time planning (economic dispatch)
![Page 62: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/62.jpg)
The timing of decisions
The day-ahead unit commitment problemMidnight
Noon
Midnight Midnight Midnight
Noon Noon Noon
![Page 63: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/63.jpg)
The timing of decisions
Intermediate-term unit commitment problem
1:15 pm 1:45 pm
1:30
2:15 pm
1:00 pm 2:00 pm 3:00 pm
2:30 pm
![Page 64: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/64.jpg)
The timing of decisions
Intermediate-term unit commitment problem
1:15 pm 1:45 pm
1:30
2:15 pm
1:00 pm 2:00 pm 3:00 pm
2:30 pm
![Page 65: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/65.jpg)
The timing of decisions
Intermediate-term unit commitment problem
Turbine 3
Turbine 22
Turbine 1
Notification time
1:15 pm
1:30
2:00 pm 3:00 pm
Ramping, but no on/off decisions.
Commitment
decisions
![Page 66: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/66.jpg)
The timing of decisions
Real-time economic dispatch problem
1pm 2pm
1:05 1:10 1:15 1:20 1:25 1:30
![Page 67: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/67.jpg)
The timing of decisions
Real-time economic dispatch problem
1pm 2pm
1:05 1:10 1:15 1:20 1:25 1:30
Run economic dispatch to perform 5 minute ramping
Run AC power flow model
![Page 68: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/68.jpg)
The timing of decisions
Real-time economic dispatch problem
1pm 2pm
1:05 1:10 1:15 1:20 1:25 1:30
Run economic dispatch to perform 5 minute ramping
Run AC power flow model
![Page 69: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/69.jpg)
The timing of decisions
Real-time economic dispatch problem
1pm 2pm
1:05 1:10 1:15 1:20 1:25 1:30
Run economic dispatch to perform 5 minute ramping
Run AC power flow model
![Page 70: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/70.jpg)
© 2010 Warren B. Powell Slide 70
Lecture outline
Simulating the PJM grid with SMART-ISO
The PJM planning process Model validation Modeling wind Integrating offshore wind
![Page 71: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/71.jpg)
SMART-ISO: Calibration
Historical generation mix during 22-28 Jul 2010
Nuclear
Steam
Comb. cycle+gasPumped hydro
![Page 72: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/72.jpg)
SMART-ISO: Calibration
Simulated generation mix during 22-28 Jul 2010
Steam
Nuclear
Comb. cycle+gasPumped hydro
![Page 73: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/73.jpg)
Actual vs. simulated LMPs
January April
July October
![Page 74: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/74.jpg)
© 2010 Warren B. Powell Slide 74
Lecture outline
Simulating the PJM grid with SMART-ISO
The PJM planning process Model validation Modeling wind Integrating offshore wind
![Page 75: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/75.jpg)
Energy from wind
30 days
Wind power from all PJM wind farms
![Page 76: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/76.jpg)
Energy from wind
Illustration of forecasted wind power and actual» The forecast (black line) is deterministic (at time t, when the forecast
was made). The actuals are stochastic.
This is our forecast of the wind power at time t’, made at time t.
'ttf
This is the actual energy from wind, showingthe deviations from forecast.
![Page 77: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/77.jpg)
Energy from wind
Two types of uncertainty arise in forecasting:» At time t, we have forecasts for different times into the
future:
• The forecast is an imperfect estimate of the actual load at time t’:
» As new information arrives, the forecasts themselves change from time t to t+1:
'ttf
1, ' , ' 1, 'f
t t t t t tf f This change in the forecast is “stochastic” at time t.
' , ' 'L
t t t tL f The actual load at time t’ is “stochastic” at time t.
![Page 78: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/78.jpg)
Forecasting wind
Rolling 24-hour forecast of PJM wind farms
Actual
Hour of day1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
Meg
awat
ts
![Page 79: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/79.jpg)
Onshore & offshore wind farms
We were given access to data on the wind power generated by onshore wind farms within PJM
Proposal: Use onshore data to calibrate a stochastic model of forecasting errors. Then use this model to create a simulated “actual” for offshore.
![Page 80: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/80.jpg)
Distribution of forecast errors» Uses adjusted spatial correlations to improve fit.
Simulating onshore wind
Observed
Simulated
Error in forecasted wind speed
![Page 81: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/81.jpg)
Simulating onshore wind Cumulative histogram of the # of consecutive time intervals the
observed/simulated time series is above the forecasted one (chosen farm only):
Time actual is above forecast
ObservedSimulated
![Page 82: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/82.jpg)
Simulating onshore wind Cumulative histogram of the # of consecutive time intervals the
observed/simulated time series is below the forecasted one (chosen farm only):
Time actual is below forecast
ObservedSimulated
![Page 83: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/83.jpg)
Wind forecast samples80
70
60
50
40
30
20
10
0
![Page 84: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/84.jpg)
Wind forecast samples80
70
60
50
40
30
20
10
0
![Page 85: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/85.jpg)
Wind forecast samples80
70
60
50
40
30
20
10
0
![Page 86: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/86.jpg)
Wind forecast samples80
70
60
50
40
30
20
10
0
![Page 87: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/87.jpg)
Simulating offshore wind
Offshore wind – Buildout level 5
Forecasted wind
![Page 88: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/88.jpg)
© 2010 Warren B. Powell Slide 88
Lecture outline
Simulating the PJM grid with SMART-ISO
The PJM planning process Model validation Modeling wind Integrating offshore wind
![Page 89: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/89.jpg)
![Page 90: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/90.jpg)
SMART-ISO: Offshore wind study
Mid-Atlantic Offshore Wind Integration and Transmission Study (U. Delaware & partners, funded by DOE)
29 offshore sub-blocks in 5 build-out scenarios:» 1: 8 GW» 2: 28 GW» 3: 40 GW» 4: 55 GW» 5: 78 GW
![Page 91: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/91.jpg)
![Page 92: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/92.jpg)
Modeling wind
» Steadier than onshore? Where???
GW
![Page 93: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/93.jpg)
Modeling wind
The power from wind:
» The cubic relationship means small changes in speed translate to large changes in power.
31
2P B Av
Wind speed (in m/sec)v
3Area of rotor blades in mA
3Density of air ( 1.225kg/m )
Power coefficient
fraction of wind converted to mechanical energy
.593 (the Betz limit)
B
![Page 94: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/94.jpg)
Unit commitment under uncertainty
Actual wind
Hour ahead
forecast
How forecasting uncertainty causes outages.
![Page 95: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/95.jpg)
SMART-ISO: Offshore wind study
![Page 96: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/96.jpg)
SMART-ISO: Offshore wind study
Less steamUniform increase in gas
![Page 97: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/97.jpg)
SMART-ISO: Offshore wind study
Outage probabilities over 21 scenarios for January, April and October:
Per
cent
of
sam
ples
ther
e is
an
outa
ge
Bas
e P
JM r
eser
veO
ptim
ized
res
erve
sP
erfe
ct in
form
atio
n
![Page 98: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/98.jpg)
SMART-ISO: Offshore wind study
Outage probabilities over 21 scenarios for July
Per
cent
of
sam
ples
ther
e is
an
outa
ge
![Page 99: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/99.jpg)
SMART-ISO: Offshore wind study
Ramping reserves, July, 2010
Perfect forecast
Imperfect forecast
16 14 12 10 8
6
4
2
0
1 2 3 4 5
Buildout levels
Ram
ping
res
erve
s (G
W)
![Page 100: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/100.jpg)
SMART-ISO: Offshore wind study
Observations» The requirement of reserves at 20 percent of capacity is only for
July, and we believe this over-estimates the reserves needed.» Other months require reserves at 10 percent of capacity, which is
still substantial, considering that renewables generate energy at roughly 20 percent of capacity.• 10 percent reserves means that 100 MW of wind generation
requires 10 MW of spinning reserve.• 100 MW of generating capacity translates to around 25 MW of
power (on average). So 25 MW of power requires 10 MW of spinning reserve from a fossil plant.
» This is for an “as is” network – we are using existing generation technologies and existing planning procedures.
» A richer portfolio including demand response, battery storage and more sophisticated planning should help.
![Page 101: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/101.jpg)
Outline
The renewables challenge Combining wind and solar The uncertainties of energy Three energy problems
» An energy storage problem» An energy portfolio policy model» Simulating the PJM grid using SMART-ISO
Concluding remarks
![Page 102: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/102.jpg)
The challenge of renewables
What did we learn?» Wind and solar are variable, and uncertain, with very
different characteristics.» Back-of-the-envelope analysis (e.g. capacity factor
analysis) completely ignores the challenges of dealing with variability and uncertainty.
» It is important to design effective policies for dealing with variability, forecasts, and uncertainty.
» Reserves to handle uncertainty can be significant, and are easily overlooked.
» Renewables are a powerful alternative to reduce our CO2 footprint, but they need to be planned properly to avoid unexpected costs.
![Page 103: Slide 1 The Renewables Challenge: Keeping the lights on while managing variability and uncertainty Princeton University Academic Mini-Reunion October 2,](https://reader035.vdocuments.mx/reader035/viewer/2022070410/56649f0c5503460f94c1fc32/html5/thumbnails/103.jpg)