renewable energy optimazation
Post on 14-Apr-2018
215 Views
Preview:
TRANSCRIPT
-
7/27/2019 renewable energy optimazation
1/35
Andy Walker PhD PE
Principal Engineer
National Renewable Energy Laboratory
Andy_Walker@nrel.gov
Acknowledgement: US DOE Federal Energy ManagementProgram
Renewable Energy Optimization
mailto:Andy_Walker@nrel.govmailto:Andy_Walker@nrel.gov -
7/27/2019 renewable energy optimazation
2/35
2
Combining Renewable Energy MeasuresPhotovoltaics
Daylighting
Biomass Heat/PowerConcentrating Solar
Heat/PowerSolar Vent Air Preheat
Solar Water HeatingWind Power
Ground Source HeatPump
Landfill Gas
-
7/27/2019 renewable energy optimazation
3/35
3
"Everything should be made assimple as possible,
but not simpler."
Albert Einstein
-
7/27/2019 renewable energy optimazation
4/35
4
Consider Even One RE Generator
E from utility (kWh) = E load (kWh) E RE generator(kWh)
kWh load
kWh renewables
kWh from utility
-
7/27/2019 renewable energy optimazation
5/35
5
Two Dimensions, power and time:E
from utility(kWh)=P
load(kW) *{t
load-t
re generator}(hours)
Eto utility(kWh)={Pre generatorPload}(kW)*tregenerator(hours)
t loadt re gen
kW load kW renewables
-
7/27/2019 renewable energy optimazation
6/35
6
Two approaches to integration:
Time Series: Identify state of system at eachtime step, and step through time series (8,760hours/year) to perform integration.
Eg. Hourly Simulations: HOMER, SAM, IMBY,PVWatts, etc.
Polynomial Expansion: Identify states that
system could be in and calculate percentage oftime system is in that state. Eg. REO. Timeperiod is arbitrary (Hourly, Monthly, Annual)
-
7/27/2019 renewable energy optimazation
7/35
7
Your choice: An approximate solution to an exact equation or
An exact solution to an approximate equation.
There is no right answer, only models. Even the performance of the completed
project is only a model of the ideal, not theright answer.
With perfect out of the question, how good isgood enough???
-
7/27/2019 renewable energy optimazation
8/35
8
Whats wrong with hourly simulation?
Lack of hourly resource data Often modeled, not measured (183/239 sites in NSRDB) Often not for the specific site High uncertainty in data
Lack of hourly load data
Not often recorded Often modeled or stipulated
Ever-increasing detail of simulation programs Requires detailed design early in evaluation of a project Often run with default values Expensive
Inaccessible in early planning
-
7/27/2019 renewable energy optimazation
9/35
9
Wind Energy in vicinity of Fairf ield CA
Project Site
Hourly Data Site
-
7/27/2019 renewable energy optimazation
10/35
10
Hourly vs Monthly Solar Radiation
Global Solar Radiation for Albuquerque NM, User's Manualfor TMY2s Typical Meteorological Years Derived from the1961-1990 National Solar Radiation Data Base, WilliamMarion and Ken Urban, June 1995
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
0 6 12 18 24
H o u r o f D ay
Standard
D
eviation(
%
)
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
0 6 12
M o n t h o f Y e a r
Standard
D
eviation
(%
)
-
7/27/2019 renewable energy optimazation
11/35
11
HOMERHourlyLoad
Inputs
-
7/27/2019 renewable energy optimazation
12/35
12
Integration of Renewables by methodof Polynomial Expansion or Mind
your Ps and Qs
P=the fraction of time the system is in a certainstate
Any number of technologiesAny number of states per technology
States that RE generator may be in:(p+q+r+s+.)=1
States that Load may be in:
(l+m+n+.)=1
-
7/27/2019 renewable energy optimazation
13/35
13
020000
40000
60000
80000
100000
120000
0 2000 4000 6000 8000
Hours
ACPower(W)
ON 66 kW for 3063 hours
OFF 0 kW for 5696 hou
Hourly from PVWatts for 100 kW, 1-axis tracking, Boulder CO
Consider two states (p=fraction time on, q=fraction time off) over time period T(p + q) = 1For seven RE generators:1*1*1*1*1*1*1=1(p1+q1)*(p2+q2)*(p3+q3)*(p4+q4)*(p5+q5)*(p6+q6)*(p7+q7)=1
-
7/27/2019 renewable energy optimazation
14/35
14
(p7*p6*p5*p4*p3*p2*p1)+(p7*p6*p5*p4*p3*p2*q1)+(p7*p6*p5*p4*p3*q2*p1)+(p7*p6*p5*p4*p3*q2*q1)+(p7*p6*p5*p4*q3*p2*p1)+(p7*p6*p5*p4*q3*p2*q1)+(p7*p6*p5*p4*q3*q2*p1)+(p7*p6*p5*p4*q3*q2*q1)+(p7*p6*p5*q4*p3*p2*p1)+(p7*p6*p5*q4*p3*p2*q1)+(p7*p6*p5*q4*p3*q2*p1)+(p7*p6*p5*q4*p3*q2*q1)+(p7*p6*p5*q4*q3*p2*p1)+(p7*p6*p5*q4*q3*p2*q1)+(p7*p6*p5*q4*q3*q2*p1)+(p7*p6*p5*q4*q3*q2*q1)+(p7*p6*q5*p4*p3*p2*p1)+(p7*p6*q5*p4*p3*p2*q1)+(p7*p6*q5*p4*p3*
q2*p1)+(p7*p6*q5*p4*p3*q2*q1)+(p7*p6*q5*p4*q3*p2*p1)+(p7*p6*q5*p4*q3*p2*q1)+(p7*p6*q5*p4*q3*q2*p1)+(p7*p6*q5*p4*q3*q2*q1)+(p7*p6*q5*q4*p3*p2*p1)+(p7*p6*q5*q4*p3*p2*q1)+(p7*p6*q5*q4*p3*q2*p1)+(p7*p6*q5*q4*p3*q2*q1)+(p7*p6*q5*q4*q3*p2*p1)+(p7*p6*q5*q4*q3*p2*q1)+(p7*p6*q5*q4*q3*q2*p1)+(p7*p6*q5*q4*q3*q2*q1)+(p7*q6*p5*p4*p3*p2*p1)+(p7*q6*p5*p4*p3*p2*q1)+(p7*q6*p5*p4*p3*q2*p1)+(p7*q6*p5*p4*p3*q2*q1)+(p7*q6*p5*p4*q3*p2*p1)+(p7*q6*p5*p4*q3*p2*q1)+(p7*q6*p5*p4*q3*q2*p1)+(p7*q6*p5*p4*q3*q2*q1)+(p7*q6*p5*q4*p3*p2*p1)+(p7*q6*p5*q4*p3*p2*q1)+(p7*q6*p5*q4*p3*q2*p1)+(p7*q6*p5*q4*p3*q2*q1)+(p7*q6*p5*q4*q3*p2*p1)+(p7*q6*p5*q4*q3*p2*q1)+(p7*q6*p5*q4*q3*q2*p1)+(p7*q6*p5*q4*q3*q2*q1)+(p7*q6*q5*p4*p3*p2*p1)+(p7*q6*q5*p4*p3*p2*q1)+(p7*q6*q5*p4*p3*q2*p1)+(p7*q6*q5*p4*p3*q2*q1)+(p7*q6*q5*p4*q3*p2*p1)+(p7*q6*q5*p4*q3*p2*q1)+(p7*q6*q5*p4*q3*q2*p1)+(p7*q6*q5*p4*q3*q2*q1)+(p7*q6*q5*q4*p3*p2*p1)+(p7*q6*q5*q4*p3*p2*q1)+(p7*q6*q5*q4*p3*q2*p1)+(p7*q6*q5*q4*p3*q2*q1)+(p7*q6*q5*q4*q3*p2*p1)+(p7*q6*q5*q4*q3*p2*q1)+(p7*q6*q5*q4*q3*q2*p1)+(p7*q6*q5*q4*q3*q2*q1)+(q7*p6*p5*p4*p3*p2*p1)+(q7*p6*p5*p4*p3*p2*q1)+(q7*p6*p5*p4*p3*q2*p1)+(q7*p6*p5*p4*p3*q2*q1)+(q7*p6*p5*p4*q3*p2*p1)+(q7*p6*p5*p4*q3*p2*q1)+(q7*p6*p5*p4*q3*q2*p1)+(q7*p6*p5*p4*q3*q2*q1)+(q7*p6*p5*q4*p3*p2*p1)+(q7*p6*p5*q4*p3*p2*q1)+(q7*p6*p5*q4*p3*q2*p1)+(q7*p6*p5*q4*p3*q2*q1)+(q7*p6*p5*q4*q3*p2*p1)+(q7*p6*p5*q4*q3*p2*q1)+(q7*p6*p5*q4*q3*q2*p1)+(q7*p6*p5*q4*q3*q2*q1)+(q7*p6*q5*p4*p3*p2*p1)+(q7*p6*q5*p4*p3*p2*q1)+(q7*p6*q5*p4*p3*q2*p1)+(q7*p6*q5*p4*p3*q2*q1)+(q7*p6*q5*p4*q3*p2*p1)+(q7*p6*q5*p4*q3*p2*q1)+(q7*p6*q5*p4*q3*q2*p1)+(q7*p6*q5*p4*q3*q2*q1)+(q7*p6*q5*q4*p3*p2*p1)+(q7*p6*q5*q4*p3*p2*q1)+(q7*p6*q5*q4*p3*q2*p1)+(q7*p6*q5*q4*p3*q2*q1)+(q7*p6*q5*q4*q3*p2*p1)+(q7*p6*q5*q4*q3*p2*q1)+(q7*p6*q5*q4*q3*q2*p1)+(q7*p6*q5*q4*q3*q2*q1)+(q7*q6*p5*p4*p3*p2*p1)+(q
7*q6*p5*p4*p3*p2*q1)+(q7*q6*p5*p4*p3*q2*p1)+(q7*q6*p5*p4*p3*q2*q1)+(q7*q6*p5*p4*q3*p2*p1)+(q7*q6*p5*p4*q3*p2*q1)+(q7*q6*p5*p4*q3*q2*p1)+(q7*q6*p5*p4*q3*q2*q1)+(q7*q6*p5*q4*p3*p2*p1)+(q7*q6*p5*q4*p3*p2*q1)+(q7*q6*p5*q4*p3*q2*p1)+(q7*q6*p5*q4*p3*q2*q1)+(q7*q6*p5*q4*q3*p2*p1)+(q7*q6*p5*q4*q3*p2*q1)+(q7*q6*p5*q4*q3*q2*p1)+(q7*q6*p5*q4*q3*q2*q1)+(q7*q6*q5*p4*p3*p2*p1)+(q7*q6*q5*p4*p3*p2*q1)+(q7*q6*q5*p4*p3*q2*p1)+(q7*q6*q5*p4*p3*q2*q1)+(q7*q6*q5*p4*q3*p2*p1)+(q7*q6*q5*p4*q3*p2*q1)+(q7*q6*q5*p4*q3*q2*p1)+(q7*q6*q5*p4*q3*q2*q1)+(q7*q6*q5*q4*p3*p2*p1)+(q7*q6*q5*q4*p3*p2*q1)+(q7*q6*q5*q4*p3*q2*p1)+(q7*q6*q5*q4*p3*q2*q1)+(q7*q6*q5*q4*q3*p2*p1)+(q7*q6*q5*q4*q3*p2*q1)+(q7*q6*q5*q4*q3*q2*p1)+(q7*q6*q5*q4*q3*q2*q1) =1
-
7/27/2019 renewable energy optimazation
15/35
15
Stochastic Integration of Renewable Energy Technologies by themethod of Polynomial Expansion (SIRET)
This simplified figure shows seven possible states the system could be in,but the model actually has 128 states for seven technologies.
100%
-
7/27/2019 renewable energy optimazation
16/35
16
What is the best integration period?
Pros Cons
Annual
(year/1)
Resource and Load dataavailable.
No seasonal or diurnalvariation
Hourly
(Year/8760)
Seasonal and DiurnalVariation
Resource and Load variableand/or unavailable
Monthly
(Year/12)
Seasonal variation
Load from Utility Bills
Resource more certain
No diurnal variation
Monthly Day/Night
(Year/24)
Load information available
Resource more certain
Seasonal Variation
Diurnal Variation
Utility bill may not tabulateby time-of-use.
-
7/27/2019 renewable energy optimazation
17/35
17
Best Mix of Renewable Energy
Technologies Depends on: Renewable Energy Resources
Technology Characterization Cost ($/kW installed, O&M Cost)
Performance (efficiency)
Economic Parameters Discount rates
Fuel Escalation Rates
State, Utility and Federal Incentives
Mandates (Executive Order, Legislation)
-
7/27/2019 renewable energy optimazation
18/35
18
Technology Characteristics
Heuristic ModelsCost
(Size, m2)*(Unit Cost, $/m2)
Performance
(Size, m2)*(Resource, kWh/m2)*(efficiency)
-
7/27/2019 renewable energy optimazation
19/35
19
Comparison of TEAM REO and Site Visi t Reports for DOE Sites
-
7/27/2019 renewable energy optimazation
20/35
20
Optimization Procedure
Site Data
Geographical Information System (GIS) Data
Incentive Data from DSIREUSA.ORG
PV SVPWind DaylightingSWH CSP Biomass
Dispatch AlgorithmDispatch Algorithm
Life Cycle CostLife Cycle Cost
Technology Characteristics.
OptimizationOptimization
-
7/27/2019 renewable energy optimazation
21/35
21
Optimization Problem
Determine the least cost
combination of renewable
energy technologies for a
facility
Objective: Minimize Life CycleCost ($)
Variables: Size of EachTechnology (kW of PV, kW ofwind, etc)
Constraints: such as 15% ofenergy from renewables
-
7/27/2019 renewable energy optimazation
22/35
22
$10,000
$100,000
$1,000,000
$10,000,000
$10,000 $100,000 $1,000,000 $10,000,000
Hourly Simulation REOCO Commercial $655,076 $614,862CO Industrial $5,158,624 $5,870,856CO Residential $62,117 $49,463WA Commercial $460,039 $568,455
WA Industrial $3,149,595 $4,533,277WA Residential $53,587 $62,304
AZ Commercial $555,003 $522,448AZ Industrial $4,469,720 $4,891,129AZ Residential $53,567 $41,256
Compare/Contrast with Hourly
Simulation
Comparison of Annual Average and Hourly Simu lation in Renewable Energy Technology System SizingChristine L. Lee, University of Colorado at Boulder, 2009
REO
Simulation
Life Cycle Cost ($)
-
7/27/2019 renewable energy optimazation
23/35
23
Compare/Contrast with Hourly
Simulation
Hourly Simulation REO % Difference
PV Energy, kWh 705,725 740,583 4.9%
Wind Energy, kWh 324,069 291,283 -10.1%Energy from Utility, kWh 1,169,264 1,407,744 20.4%Energy to Utility, kWh 139,478 380,030 172.5%Life Cycle Cost, $ $5,158,624 $5,870,856 13.8%
Percent Renewables 50% 41.80% -16.4%
Comparison of Annual Average and Hourly Simu lation in Renewable Energy Technology System SizingChristine L. Lee, University of Colorado at Boulder, 2009
Industrial Load Profile; Colorado Climate; 2,059,580 kWh/year load;464 kW PV; 493 kW Wind
-
7/27/2019 renewable energy optimazation
24/35
24
Examples of Renewable Energy
Optimization (REO) Analysis Town of Greensburg, KS Frito Lay North America plants (7) National Zoo, DC Anheuser Busch facilities (62) High School in Sun Valley, ID
San Nicolas Island, CA DOE Waste Isolation Pilot Plant, NM DOE Savannah River Plant SC Agricultural Research Stations in TX (8) DOE Laboratories (31) Air Force Bases (85) GSA Land Ports of Entry (121) DHS Land Ports of Entry (32) USCG Bases (3) Presidio of San Francisco Others.
-
7/27/2019 renewable energy optimazation
25/35
25
Results of Renewable Energy Optimization:Technology Sizes
PhotovoltaicsSize (kW)
WindCapacity(kW)
Solar VentPreheatArea (ft2)
SolarWaterHeatingArea (ft2)
SolarThermalArea (ft2)
SolarThermalElectric(kW)
Biomass BoilerSize (M Btu/h)
BiomassCogenerationSize (kW)
DaylightingOffice UtilitySkylight/FloorArea Ratio
DaylightingWarehouseSkylight/FloorArea Ratio
10 0 15258 29,179 0 0 0 0 3.968% 3.036%0 957 0 0 0.00 0.00 1.34 101 0.000% 0.000%
Example: Town of Greensburg KS
BuildingsCentral Plant
-
7/27/2019 renewable energy optimazation
26/35
26
Annual Energy from Each Technology
(with Basecase)
0
20000
40000
60000
80000
100000
120000
Base
case
RE
Case
An
nualEnergy(Mbtu)
Electric (Mbtu) Natural Gas (Mbtu) Other Fuel (Mbtu)
Photovoltaics (Mbtu) Wind (Mbtu) Solar Vent Preheat (Mbtu)
Solar Water Heating Solar Themal (Mbtu) Biomass (Mbtu)
Daylighting (Mbtu)
Example: Town of Greensburg KS
-
7/27/2019 renewable energy optimazation
27/35
27
Wind Energy
Build ing Name Wind Capaci ty (kW)
Wind Annual
EnergyDelivery
(kWh/year)
Capacity
Factor (%)
WindAnnual Cost
Savings ($)
Wind
AnnualO&M Cost
($/year)
Wind
PaybackPeriod
(years)Central Plant 957 2,533,894 30.23% $276,194 $7,560 7.0
-
7/27/2019 renewable energy optimazation
28/35
28
Example: Frito Lay North AmericaMinimum Life Cycle Cost (no constraints)
0
200000
400000
600000
800000
1000000
1200000
Plant#
1Basecase
Plant#
1RE
Case
Plant#
2Basecase
Plant
#2RE
Case
Plant#
3Ba
secase
Plant
#3RE
Case
Plant#
4BaseC
ase
Plant#
4RE
Case
Plant$
5Ba
secase
Plant
#5RE
Case
Plant
#6Basecas
e
Plant
#6RE
Case
Plant#
7Ba
secase
Plant
#7RE
Case
AnnualEnergy(Mbtu)
Electric (Mbtu) Natural Gas (Mbtu) Other Fuel (Mbtu)
Photovoltaics (Mbtu) Wind (Mbtu) Solar Vent Preheat (Mbtu)
Solar Themal (Mbtu) Biomass (Mbtu) Daylighting (Mbtu)
-
7/27/2019 renewable energy optimazation
29/35
29
Frito Lay North AmericaSolar Thermal at Sunchips Plant
Modesto CA
-
7/27/2019 renewable energy optimazation
30/35
30
Example: Frito Lay North AmericaMinimum Life Cycle Cost (Net Zero constraint)
0
200000
400000
600000
800000
1000000
1200000
Plant
#1Ba
secase
Plant
#1Net
Zero
Plant
#2Ba
secase
Plant
#2Net
Zero
Plant
#3Basecase
Plant
#3Net
Zero
Plant
#4BaseC
ase
Plant
#4Net
Zero
Plant
#5Basecase
Plant#
5Net
Zero
Plant
#6Ba
secase
Plant#
6Net
Zero
Plant#
7
Plant
#7Net
Zero
AnnualEnergy(Mbtu)
Electric (Mbtu) Natural Gas (Mbtu) Other Fuel (Mbtu)
Photovoltaics (Mbtu) Wind (Mbtu) Solar Vent Preheat (Mbtu)
Solar Themal (Mbtu) Biomass (Mbtu) Daylighting (Mbtu)
-
7/27/2019 renewable energy optimazation
31/35
31
Example: Frito Lay Optimization
for Seven Manufacturing PlantsConstraint: Net Zero
Photovoltaics
Size (kW)
Wind
Capacity
(kW)
Solar Vent
Preheat
Area (ft 2)
Solar
Thermal
Area (f t2)
Biomass
Boiler Size
(M Btu/h)
Biomass
Cogeneration
Size (kW)
Daylighting
Office Utility
Skylight/Floor
Area Ratio
Daylighting
Warehouse
Skylight/Floor
Area RatioPlant #1 200 491 5456 509196 19 1669 2.2% 2.1%
Plant #2 0 6187 8953 391987 87 3097 3.8% 2.0%
Plant #3 0 3107 13098 469621 44 3180 4.9% 3.6%
Plant #4 1011 1000 10213 1360535 78 4108 3.4% 1.8%
Plant #5 1003 998 10327 704140 44 3327 6.1% 3.4%
Plant #6 0 0 10322 1529609 74 6020 err err
Plant #7 0 3699 10802 673761 43 2193 3.3% 3.7%
-
7/27/2019 renewable energy optimazation
32/35
32
-
7/27/2019 renewable energy optimazation
33/35
33
REO Example: Net Zero ZooNational Zoological Park (NZP) and Conservation Research Center
(CRC), Washington DC
-40000
-20000
0
20000
40000
60000
80000
100000
120000
140000
160000
Base
case
RECase
AnnualEnergy(M
btu)
Daylighting (Mbtu)
Biomass (Mbtu)
Solar Themal (Mbtu)
Solar Water Heating
Solar Vent Preheat (Mbtu)
Wind (Mbtu)
Photovoltaics (Mbtu)
Other Fuel (Mbtu)
Natural Gas (Mbtu)
Electric (Mbtu)
Electric
generation
at CRC
Cancels
remaining
gas use at
NZP
Zoo EntranceTai Shan the Panda
-
7/27/2019 renewable energy optimazation
34/35
34
Initial Cost for Renewable Energy Projects ($) $269,634,839Annual Electric Savings (kWh/year) 217,734,776
Annual Gas/Fuel Savings (therms/year) 8,693,565
Annual Cost Savings ($/year) $26,241,533
Simple Payback Period (years) 10.3Rate of Return 11.3%
Sizes of Each Technology:Photovoltaics (kW) 0Wind Energy (kW) 41714Solar Ventilation Air Preheat (sf) 460988
Solar Water Heating (sf) 1274264Solar Thermal Parabolic Trough (sf) 452991
Solar Thermal Electric (kW) 0Biomass Gasification Boiler (MBH) 51Biomass Gasification Cogen (kW) 139
Biomass Anaerobic Digester r (Ft3)) 0Biomass Anaerobic Digester Cogen (kW) 0
Skylight Area (sf) 2602033
31 DOESites
050000
100000150000
200000250000300000350000400000
AmesLaborat
ory
Argonn
eNatio
nalLaboratorySiteD
BrookhavenNationalLaboratory
FermiN
ationalA
ccelerat
orLaboratory
Forre
stalH
eadquarters
Germ
antow
n
IdahoN
ation
alLa
boratory -I
daho
Falls
IdahoNationa
lLaboratory
-Scoville
Knolls
Atomic -
Kesserling
KnollsAtom
ic -Niskayuna
LasVegas
NV/L
V
Lawrence
Berkele
yNationalLaborat
ory
...LawrenceLiverm
oreNationalLabora
LosAlamo
sNationalLaboratory
NETL
Albany
NETL
Morgantown
Nevada
TestSite
OakRidg
eNation
alLabO
ffice
(Oak
Ridg
eNati
onalLaboratory (X
-10
PacificNorthwes
tNationalLaboratory
PaducahG
aseous
PantexSiteOffice
Princ
etonPlasm
aPhysicsLaboratory
Richlan
dOperatio
nsOffic
e -Hanford1
SandiaN
ationalLaboratoryCA
SandiaNa
tionalLaboratoryNM
SandiaN
ation
alLaboratoryNV
Savannah
River
StanfordLine
arAcceleratorCenter
..Thoma
sJefferso
nNationalAc
celerat
WasteIsolation
PilotPlant
Y-12
AnnualEn
ergy(Mbtu)
Wind (Mbtu) Solar Vent Preheat (Mbtu) Solar Water Heating
Solar Thermal (Mbtu) Biomass Gasifier (Mbtu) Anaerobic Digester (Mbtu)
Daylighting (Mbtu) Photovoltaics (Mbtu)
-
7/27/2019 renewable energy optimazation
35/35
35
It's so much easier tosuggest solutions
when you don't knowtoo much about the
problem.
- Malcolm Forbes
Thank You!
Andy WalkerSenior Engineer
National Renewable Energy LaboratoryAndy_Walker@nrel.gov
Simplicity is quiteoften the mark ofexcellence-Allen Bennett, SMDC
mailto:Andy_Walker@nrel.govmailto:Andy_Walker@nrel.gov
top related