Leveraging Plug-In Hybrids to Increase Solar Deployment in California:
Leveraging Plug-In Hybrids to Increase Solar Deployment in California: The Economic Effect of Plug-in Hybrids on Distributed Photovoltaic Systems
Robert Broesler, Materials Science and Engineering David Jacobowitz, Goldman School of Public Policy Nate Miller, Materials Science and Engineering Daniel Prull, Renewable and Appropriate Energy
Laboratory
1
Big Picture Project Motivation: ¾ Utilities have special EV TOU and net metering rates ¾ Can a residential plug-in hybrid vehicle (PHEV) owner take advantage of these rates to increase the value of a new PV installation?
Our Approach: ¾ For each major utility in CA, model the rate structures, insolation and demand profiles ¾ Determine payback time and NPV of a PV system with and without a PHEV 2
Data and Assumptions ¾ Real hourly average residential load data for 2001 for Northern and Southern California (PG&E and SCE respectively) ¾ Real hourly insolation data for 2001 from California Irrigation Management Information Systems ¾ Current TOU, EV-TOU and Normal Rate structures from PG&E, SCE, SDG&E, SMUD ¾ Empirical size dependant PV system cost (from 19982005) ¾ Assume Ideal Night PHEV Charging
3
Rates ¾E- Rates are complicated ¾Vary by…
¾Utility (PG&E, SCE, SDG&E) ¾“schedule” (Basic Residential, Time-of-Use,
various experimental scheds.) ¾Baseline Allocation (how much power do your neighbors use?) ¾Actual Use (“tiering”) ¾Time of day, day of week, season, holidays ¾Astrological Sign (no, not really) 4
Rates & Net Metering ¾Net Metering makes it yet more complicated. ¾What fraction of generated vs used power goes into which period, and which tier? ¾Reconstructing bill is not for the fainthearted
5
An example... PG&E PHEV-TOU
Summer
Winter
Tier 1
Tier 2
Tier 3
Tier 4
Tier 5
Off
0.04965
0.04965
0.14962
0.24203
0.29046
Part-Peak
0.10395
0.10395
0.20392
0.29633
0.34476
Peak
0.28368
0.28368
0.28365
0.47606
0.52449
Off
0.05795
0.05795
0.15792
0.25033
0.29876
Part-Peak
0.10383
0.10383
0.2038
0.29621
0.34464
Tier 1
Tier 2
Tier 3
Tier 4
Tier 5
Off
0.09422
0.10981
0.20978
0.30219
0.35062
Part-Peak
0.1178
0.12737
0.22734
0.31975
0.36818
Peak
0.20861
0.2241
0.32417
0.41658
0.46501
Off
0.10046
0.11605
0.21602
0.30843
0.35686
Part-Peak
0.12417
0.13976
0.23973
0.33214
0.38057
TOU
Summer
Winter
Calculation of a given bill requires the use of most elements of this matrix. 6
Model Results
PV Payback by Utility and Demand ¾PG&E and SCE are the best for PV Payback ¾Solar is almost never economic for SDG&E ¾SMUD is left off because it is never economic PV + PHEV vs. PHEV
PV vs. No PV
Optimal Pay Back Optimal Pay Back PV Size Maximum Time PV Size Maximum Time (W) NPV ($) (years) (W) NPV ($) (years) Utility Demand 1.0x 1020 $2,512 18 1020 -$1,516 NA PGE 1.6x 3420 $14,650 16 2940 $9,381 17.5 2.1x 5520 $29,430 14.5 4680 $24,820 14.5 1.0x 2880 $3,729 21 2070 -$849 NA SCE 1.6x 4440 $6,487 21 3420 $5,864 20 2.1x 5760 $9,027 20.5 4560 $16,500 16.5 1.0x 0 $0 NA 0 $0 NA SDGE 1.6x 0 $0 NA 0 $0 NA 2.1x 0 $0 NA 1020 $1,560 25
8
Pacific Gas and Electric Pacific Gas & Electric -- 1.6x average use
Pacific Gas & Electric -- 2.1x average use
4
3
15000
x 10
2.5 2 Dollars
Dollars
10000
5000
1 0.5
0
0
PV/TOU vs NoPV/Normal PV+PHEV/PHEVTOU vs PHEV/PHEVTOU -5000 0
1.5
1000
2000
3000
4000 Watts
5000
6000
7000
8000
-0.5 0
PV/TOU vs NoPV/Normal PV+PHEV/PHEVTOU vs PHEV/PHEVTOU 1000
2000
3000
4000
5000
6000
7000
8000
Watts
9
Rates, Demand and PV Payback (PG&E)
Normalized Demand and Insolation Data for Maximum Demand Day with PGE's Normalized TOU Baseline Rate
1
0.8
0.6
Pacific Gas & Electric -- 1.6x
4
2.5
x 10
0.4
2 1.5
0.2
Insolation for summer day TOU Baseline Rates Total Demand for same day
Dollars
1 0
0.5 0
4
6
8
10
12 14 Hours of the Day
16
18
20
22
24
Normalized Demand and Insolation Data for Maximum Demand Day with PGE's Normalized PHEVTOU Baseline Rates
-0.5
1
Normal (PV+PHEV vs No PV, No PHEV)
-1
PHEVTOU (PV+PHEV vs No PV, No PHEV) -1.5 -2 0
2
0.8
TOU (PV+PHEV vs No PV, No PHEV) 1000
2000
3000
4000 Watts
5000
6000
7000
8000
0.6
0.4
0.2
0
Insolation for summer day PHEVTOU Baseline Rates Total demand for same day 2
4
6
8
10
12 14 Hours of the Day
16
18
20
22
24
10
Southern California Edison 4
1
x 10
Southern California Edison -- 1.6x average use
4
2
x 10
Southern California Edison -- 2.1x average use
1.5
0.5
1 Dollars
Dollars
0 -0.5
0.5 0
-1
-0.5 -1.5 -2 0
-1
PV/TOU vs NoPV/Normal PV+PHEV/PHEVTOU vs PHEV/PHEVTOU 1000
2000
3000
4000 Watts
5000
6000
7000
8000
-1.5 0
PV/TOU vs NoPV/Normal PV+PHEV/PHEVTOU vs PHEV/PHEVTOU 1000
2000
3000
4000 Watts
5000
6000
7000
8000
11
Rates, Demand and PV Payback (SCE)
Normalized Demand and Insolation Data for Maximum Demand Day with SCEs Normalized TOU Baseline Rates
1
0.8
0.6
Southern California Edison -- 1.6x
4
2.5
x 10
0.4
2 0.2
1.5
Dollars
1
0
Total Demand for Summer Day Insolation for same day TOU Baseline Rates 2
4
6
8
10
12 14 Hours of the Day
16
18
20
22
24
0.5 Normalized Demand and Insolation Data for the Maximum Day with SCEs Normalized PHEVTOU Baseline Rates
0 1
-0.5 Normal (PV+PHEV vs No PV, No PHEV) PHEVTOU (PV+PHEV vs No PV, No PHEV) TOU (PV+PHEV vs No PV, No PHEV)
-1 -1.5 0
1000
2000
3000
4000 Watts
5000
6000
7000
8000
0.8
0.6
0.4
0.2
0
Total demand for summer day Insolation for the same day PHEVTOU Baseline Rates 2
4
6
8
10
12 14 Hours of the Day
16
18
20
22
24
12
San Diego Gas and Electric 4
0.5
x 10
San Diego Gas & Electric -- 1.6x average use
4
0.5
0
San Diego Gas & Electric -- 2.1x average use
0
-0.5
-0.5
-1
Dollars
Dollars
x 10
-1.5
-1 -1.5
-2 -2.5 -3 0
-2
PV/TOU vs NoPV/Normal PV+PHEV/PHEVTOU vs PHEV/PHEVTOU 1000
2000
3000
4000 Watts
5000
6000
7000
8000
-2.5 0
PV/TOU vs NoPV/Normal PV+PHEV/PHEVTOU vs PHEV/PHEVTOU 1000
2000
3000
4000 Watts
5000
6000
7000
8000
13
SDG&E’s Minimal and Inverted TOU Rate Structures
Normalized Demand and Insolation Data for Maximum Day with SDGEs Normalized TOU Baseline Rates
1
0.8
0.6
0.4
0.2
0
Maximum Demand Day Insolation for the same day TOU Baseline Rates 2
4
6
8
10
12 14 Hours of the Day
16
18
20
22
24
Normalized Demand and Insolation Data for Maximum Day with SDGE's Normalized PHEVTOU Baseline Rates
1
0.8
0.6
0.4
0.2
0
Maximum Demand Day Insolation for the same day PHEVTOU Baseline Rates 2
4
6
8
10
12 14 Hours of the Day
16
18
20
22
24
14
Conclusions ¾ Benefit of PHEV for PV payback from increased demand, not rate structure ¾ Benefit decreases with increasing demand factor (pre-PHEV demand)
¾ Four beneficial rate structure characteristics: ¾ ¾ ¾ ¾
High TOU peak to off-peak price differential Aggressive tier structure TOU peak time well matched to peak insolation time High normal electricity rates
¾ PHEV rate structures optimized to shift added demand away from peak demand, which usually does not coincide with peak insolation. ¾ PHEV + PV can decrease the peak and raise the minimum for system load 15
Extras
Effect of PHEV + PV on Load ¾ Assumed 1 million PHEV charging at night and 1 million optimized PV systems (5.515.75kW) ¾ Pulls down the peak by 2.78% and increases trough by 3.66% for maximum day
17
Utility Load Data ¾ Downloaded empirical total load data for LDWP, PG&E, SCE, SDG&E, SMUD from FERC for 2001 ¾ Downloaded empirical average residential demand from PG&E and SCE for 2001. ¾ We use average PG&E demand for N. California (PG&E, SMUD) and SCE for S. California (SCE, LDWP, SDG&E)
18
California Utility Distribution ¾ Five utilities make up more than 80% of the California electricity distribution and span almost the entire state. ¾ LDWP is left out of the NPV calculations because their rates are not published ¾ SCE and PG&E make up more than 60% of CA’s total.
Company % of CA Supply SCE PG&E LDWP SDG&E SMUD Total
31% 30% 9% 7% 4% 81%
Fi ve Major Utility Service Areas: Pacific Gas & Electric Sacramento Municipal Ut ility District Southern Californ ia Edison Los Angeles Department of Water and Power San Diego Gas & Electric
19
Insolation Data: ¾ California Irrigation Management Information Systems (CIMIS) hourly data ¾ 81 active stations throughout 2001 ¾ CIMIS pyranometers measure global horizontal radiation (W/m^2)
Southern California CIMIS Station Map
20
Insolation Data: ¾ Insolation data corrected for: ¾ Latitude and Longitude at each station ¾ Location of sun in the sky throughout the year ¾ Panel tilt (facing south at latitude to make maximum power)
¾ Individual station data then averaged for each utility ¾ Maximum and minimum annual insolation also used
corrected
21
Photovoltaic System Assumptions ¾ Use empirical PV system cost data for systems installed between 1998 and 2005 to get gross system cost (Wiser, 2006). ¾ CEC rebate of $2.50/Wac and 30% federal tax credit. ¾ $777 set up costs ($277 TOU meter fee and $500 permit fee). ¾ One time inverter replacement for $700/kW at year 15. ¾ 12% system efficiency and 25 year cell lifetime. Empirical System Size Cost
Mean Pre-rebate Installed Cost (2004 $/Wac)
11 10.5 10 9.5 9
y = 10.563x-0.0781 8.5 8 0
5
10
15
20
System Size (kW)
25
30
22
A few more curveballs… ¾SDG&E TOU rates have extremely small differential. (result of adjustment from lawsuit) ¾SCE prices reflect fraction of DWR vs. URG power (varies daily) ¾Some of the “best” rates for PV are discontinued (eg PG&E E-7) ¾Daily minima, meter charges 23
Using NPV as a Metric ¾ Construct individual MATLAB models for each system parameter (rates, insolation, etc.) ¾ Combine models and calculate a customer’s yearly cost and energy use ¾ Optimize scenarios based on PV system net present value (NPV) ¾ Investigate variations in customer demand, site insolation, PV system size and PHEV charging scenarios for each utility
24
PHEV Charging Scenarios ¾ Optimal Night Charging: peaks at 2am ¾ ¾ Evening Charging: peaks at 9pm ¾ Curves smoothed to have a 1 hour standard deviation PHEV Charging Scenarios 1000
800 CAR_EVE CAR_NIGHT Watt-Hours
600
400
200
0 0
5
10
15
20
25
-200 Hour of Day
25
NPV Calculation t
t
NPV = S + Y * ∑ (1 + r ) + d * ∑ i =1
−i
i =1
( )
1+ re i 1+ r
¾S = system cost (yr zero) ¾Y = maint. costs. ¾d = savings on bill from PV sys. ¾t = 25 yrs ¾r = discount rate ¾5.3%/yr, OMB std.
¾re= growth rate for electricity cost ¾also 5.3%/yr
26
Why two discount rates? ¾ One (r) is used to calculate the nominal value of future costs/savings in year-zero dollars relative to an alternative investment ¾ The other (re) is used to model the increased nominal value of electric power over time (relative to all other goods). ¾ If r=re, then current value of the savings stream is the simple sum of yearly values. ¾ It’s difficult to predict either value! 27
Payback Time (PG&E) Plots vary by residential ‘demand factor’ ¾ Additional demand of the PHEV decreases PV payback time for customers with relatively low household electricity use ¾ The greater the customer’s base electricity use (demand factor), the more quickly the PV system pays for itself ¾ As a customer’s electricity use increases, the PHEV no longer significantly improves PV payback time
28
Payback Time (SCE) ¾ The effects of varying insolation and electricity demand do change the PV payback time ¾ These effects are minor compared to the dependence on rate structure ¾ SCE and SDG&E have non-tiered EV-TOU rate structures which negate the benefit of a PHEV on PV payback time ¾ PG&E’s rate structure increases the value of PV when a PHEV is included
29
Effect of PHEV + PV on CO2 ¾ Same 1 million PHEV and PV system scenario. ¾ 0.97 lbs of CO2 per kWh of grid charging from weighted average of CA electricity generation. ¾ 23.8 lbs of CO2 per gallon combusted gasoline. ¾ Car can drive 4 miles per 1 kWh of battery power. ¾ 20 mi of electricity only driving per day per car. ¾ Replaced car with 25 mpg fuel efficiency. ¾ Found that PHEV saves approximately 0.7 lbs of CO2 per electricity only mile driven. ¾ Scenario would save 0.003ppm per year of CO2 from entering the atmosphere.
Source Avoided Gasoline Driving PHEV Charging PV Systems Net Change
CO2 (lbs) CO2 (ppm) -6.95E+09 1.44E+09 -8.30E+09 -1.38E+10
-0.0015 3.10E-04 -0.0018 -0.003
30
Robert Broesler, Materials Science and Engineering David Jacobowitz, Goldman School of Public Policy Nate Miller, Materials Science and Engineering Daniel Prull, Renewable and Appropriate Energy
Laboratory
1
Big Picture Project Motivation: ¾ Utilities have special EV TOU and net metering rates ¾ Can a residential plug-in hybrid vehicle (PHEV) owner take advantage of these rates to increase the value of a new PV installation?
Our Approach: ¾ For each major utility in CA, model the rate structures, insolation and demand profiles ¾ Determine payback time and NPV of a PV system with and without a PHEV 2
Data and Assumptions ¾ Real hourly average residential load data for 2001 for Northern and Southern California (PG&E and SCE respectively) ¾ Real hourly insolation data for 2001 from California Irrigation Management Information Systems ¾ Current TOU, EV-TOU and Normal Rate structures from PG&E, SCE, SDG&E, SMUD ¾ Empirical size dependant PV system cost (from 19982005) ¾ Assume Ideal Night PHEV Charging
3
Rates ¾E- Rates are complicated ¾Vary by…
¾Utility (PG&E, SCE, SDG&E) ¾“schedule” (Basic Residential, Time-of-Use,
various experimental scheds.) ¾Baseline Allocation (how much power do your neighbors use?) ¾Actual Use (“tiering”) ¾Time of day, day of week, season, holidays ¾Astrological Sign (no, not really) 4
Rates & Net Metering ¾Net Metering makes it yet more complicated. ¾What fraction of generated vs used power goes into which period, and which tier? ¾Reconstructing bill is not for the fainthearted
5
An example... PG&E PHEV-TOU
Summer
Winter
Tier 1
Tier 2
Tier 3
Tier 4
Tier 5
Off
0.04965
0.04965
0.14962
0.24203
0.29046
Part-Peak
0.10395
0.10395
0.20392
0.29633
0.34476
Peak
0.28368
0.28368
0.28365
0.47606
0.52449
Off
0.05795
0.05795
0.15792
0.25033
0.29876
Part-Peak
0.10383
0.10383
0.2038
0.29621
0.34464
Tier 1
Tier 2
Tier 3
Tier 4
Tier 5
Off
0.09422
0.10981
0.20978
0.30219
0.35062
Part-Peak
0.1178
0.12737
0.22734
0.31975
0.36818
Peak
0.20861
0.2241
0.32417
0.41658
0.46501
Off
0.10046
0.11605
0.21602
0.30843
0.35686
Part-Peak
0.12417
0.13976
0.23973
0.33214
0.38057
TOU
Summer
Winter
Calculation of a given bill requires the use of most elements of this matrix. 6
Model Results
PV Payback by Utility and Demand ¾PG&E and SCE are the best for PV Payback ¾Solar is almost never economic for SDG&E ¾SMUD is left off because it is never economic PV + PHEV vs. PHEV
PV vs. No PV
Optimal Pay Back Optimal Pay Back PV Size Maximum Time PV Size Maximum Time (W) NPV ($) (years) (W) NPV ($) (years) Utility Demand 1.0x 1020 $2,512 18 1020 -$1,516 NA PGE 1.6x 3420 $14,650 16 2940 $9,381 17.5 2.1x 5520 $29,430 14.5 4680 $24,820 14.5 1.0x 2880 $3,729 21 2070 -$849 NA SCE 1.6x 4440 $6,487 21 3420 $5,864 20 2.1x 5760 $9,027 20.5 4560 $16,500 16.5 1.0x 0 $0 NA 0 $0 NA SDGE 1.6x 0 $0 NA 0 $0 NA 2.1x 0 $0 NA 1020 $1,560 25
8
Pacific Gas and Electric Pacific Gas & Electric -- 1.6x average use
Pacific Gas & Electric -- 2.1x average use
4
3
15000
x 10
2.5 2 Dollars
Dollars
10000
5000
1 0.5
0
0
PV/TOU vs NoPV/Normal PV+PHEV/PHEVTOU vs PHEV/PHEVTOU -5000 0
1.5
1000
2000
3000
4000 Watts
5000
6000
7000
8000
-0.5 0
PV/TOU vs NoPV/Normal PV+PHEV/PHEVTOU vs PHEV/PHEVTOU 1000
2000
3000
4000
5000
6000
7000
8000
Watts
9
Rates, Demand and PV Payback (PG&E)
Normalized Demand and Insolation Data for Maximum Demand Day with PGE's Normalized TOU Baseline Rate
1
0.8
0.6
Pacific Gas & Electric -- 1.6x
4
2.5
x 10
0.4
2 1.5
0.2
Insolation for summer day TOU Baseline Rates Total Demand for same day
Dollars
1 0
0.5 0
4
6
8
10
12 14 Hours of the Day
16
18
20
22
24
Normalized Demand and Insolation Data for Maximum Demand Day with PGE's Normalized PHEVTOU Baseline Rates
-0.5
1
Normal (PV+PHEV vs No PV, No PHEV)
-1
PHEVTOU (PV+PHEV vs No PV, No PHEV) -1.5 -2 0
2
0.8
TOU (PV+PHEV vs No PV, No PHEV) 1000
2000
3000
4000 Watts
5000
6000
7000
8000
0.6
0.4
0.2
0
Insolation for summer day PHEVTOU Baseline Rates Total demand for same day 2
4
6
8
10
12 14 Hours of the Day
16
18
20
22
24
10
Southern California Edison 4
1
x 10
Southern California Edison -- 1.6x average use
4
2
x 10
Southern California Edison -- 2.1x average use
1.5
0.5
1 Dollars
Dollars
0 -0.5
0.5 0
-1
-0.5 -1.5 -2 0
-1
PV/TOU vs NoPV/Normal PV+PHEV/PHEVTOU vs PHEV/PHEVTOU 1000
2000
3000
4000 Watts
5000
6000
7000
8000
-1.5 0
PV/TOU vs NoPV/Normal PV+PHEV/PHEVTOU vs PHEV/PHEVTOU 1000
2000
3000
4000 Watts
5000
6000
7000
8000
11
Rates, Demand and PV Payback (SCE)
Normalized Demand and Insolation Data for Maximum Demand Day with SCEs Normalized TOU Baseline Rates
1
0.8
0.6
Southern California Edison -- 1.6x
4
2.5
x 10
0.4
2 0.2
1.5
Dollars
1
0
Total Demand for Summer Day Insolation for same day TOU Baseline Rates 2
4
6
8
10
12 14 Hours of the Day
16
18
20
22
24
0.5 Normalized Demand and Insolation Data for the Maximum Day with SCEs Normalized PHEVTOU Baseline Rates
0 1
-0.5 Normal (PV+PHEV vs No PV, No PHEV) PHEVTOU (PV+PHEV vs No PV, No PHEV) TOU (PV+PHEV vs No PV, No PHEV)
-1 -1.5 0
1000
2000
3000
4000 Watts
5000
6000
7000
8000
0.8
0.6
0.4
0.2
0
Total demand for summer day Insolation for the same day PHEVTOU Baseline Rates 2
4
6
8
10
12 14 Hours of the Day
16
18
20
22
24
12
San Diego Gas and Electric 4
0.5
x 10
San Diego Gas & Electric -- 1.6x average use
4
0.5
0
San Diego Gas & Electric -- 2.1x average use
0
-0.5
-0.5
-1
Dollars
Dollars
x 10
-1.5
-1 -1.5
-2 -2.5 -3 0
-2
PV/TOU vs NoPV/Normal PV+PHEV/PHEVTOU vs PHEV/PHEVTOU 1000
2000
3000
4000 Watts
5000
6000
7000
8000
-2.5 0
PV/TOU vs NoPV/Normal PV+PHEV/PHEVTOU vs PHEV/PHEVTOU 1000
2000
3000
4000 Watts
5000
6000
7000
8000
13
SDG&E’s Minimal and Inverted TOU Rate Structures
Normalized Demand and Insolation Data for Maximum Day with SDGEs Normalized TOU Baseline Rates
1
0.8
0.6
0.4
0.2
0
Maximum Demand Day Insolation for the same day TOU Baseline Rates 2
4
6
8
10
12 14 Hours of the Day
16
18
20
22
24
Normalized Demand and Insolation Data for Maximum Day with SDGE's Normalized PHEVTOU Baseline Rates
1
0.8
0.6
0.4
0.2
0
Maximum Demand Day Insolation for the same day PHEVTOU Baseline Rates 2
4
6
8
10
12 14 Hours of the Day
16
18
20
22
24
14
Conclusions ¾ Benefit of PHEV for PV payback from increased demand, not rate structure ¾ Benefit decreases with increasing demand factor (pre-PHEV demand)
¾ Four beneficial rate structure characteristics: ¾ ¾ ¾ ¾
High TOU peak to off-peak price differential Aggressive tier structure TOU peak time well matched to peak insolation time High normal electricity rates
¾ PHEV rate structures optimized to shift added demand away from peak demand, which usually does not coincide with peak insolation. ¾ PHEV + PV can decrease the peak and raise the minimum for system load 15
Extras
Effect of PHEV + PV on Load ¾ Assumed 1 million PHEV charging at night and 1 million optimized PV systems (5.515.75kW) ¾ Pulls down the peak by 2.78% and increases trough by 3.66% for maximum day
17
Utility Load Data ¾ Downloaded empirical total load data for LDWP, PG&E, SCE, SDG&E, SMUD from FERC for 2001 ¾ Downloaded empirical average residential demand from PG&E and SCE for 2001. ¾ We use average PG&E demand for N. California (PG&E, SMUD) and SCE for S. California (SCE, LDWP, SDG&E)
18
California Utility Distribution ¾ Five utilities make up more than 80% of the California electricity distribution and span almost the entire state. ¾ LDWP is left out of the NPV calculations because their rates are not published ¾ SCE and PG&E make up more than 60% of CA’s total.
Company % of CA Supply SCE PG&E LDWP SDG&E SMUD Total
31% 30% 9% 7% 4% 81%
Fi ve Major Utility Service Areas: Pacific Gas & Electric Sacramento Municipal Ut ility District Southern Californ ia Edison Los Angeles Department of Water and Power San Diego Gas & Electric
19
Insolation Data: ¾ California Irrigation Management Information Systems (CIMIS) hourly data ¾ 81 active stations throughout 2001 ¾ CIMIS pyranometers measure global horizontal radiation (W/m^2)
Southern California CIMIS Station Map
20
Insolation Data: ¾ Insolation data corrected for: ¾ Latitude and Longitude at each station ¾ Location of sun in the sky throughout the year ¾ Panel tilt (facing south at latitude to make maximum power)
¾ Individual station data then averaged for each utility ¾ Maximum and minimum annual insolation also used
corrected
21
Photovoltaic System Assumptions ¾ Use empirical PV system cost data for systems installed between 1998 and 2005 to get gross system cost (Wiser, 2006). ¾ CEC rebate of $2.50/Wac and 30% federal tax credit. ¾ $777 set up costs ($277 TOU meter fee and $500 permit fee). ¾ One time inverter replacement for $700/kW at year 15. ¾ 12% system efficiency and 25 year cell lifetime. Empirical System Size Cost
Mean Pre-rebate Installed Cost (2004 $/Wac)
11 10.5 10 9.5 9
y = 10.563x-0.0781 8.5 8 0
5
10
15
20
System Size (kW)
25
30
22
A few more curveballs… ¾SDG&E TOU rates have extremely small differential. (result of adjustment from lawsuit) ¾SCE prices reflect fraction of DWR vs. URG power (varies daily) ¾Some of the “best” rates for PV are discontinued (eg PG&E E-7) ¾Daily minima, meter charges 23
Using NPV as a Metric ¾ Construct individual MATLAB models for each system parameter (rates, insolation, etc.) ¾ Combine models and calculate a customer’s yearly cost and energy use ¾ Optimize scenarios based on PV system net present value (NPV) ¾ Investigate variations in customer demand, site insolation, PV system size and PHEV charging scenarios for each utility
24
PHEV Charging Scenarios ¾ Optimal Night Charging: peaks at 2am ¾ ¾ Evening Charging: peaks at 9pm ¾ Curves smoothed to have a 1 hour standard deviation PHEV Charging Scenarios 1000
800 CAR_EVE CAR_NIGHT Watt-Hours
600
400
200
0 0
5
10
15
20
25
-200 Hour of Day
25
NPV Calculation t
t
NPV = S + Y * ∑ (1 + r ) + d * ∑ i =1
−i
i =1
( )
1+ re i 1+ r
¾S = system cost (yr zero) ¾Y = maint. costs. ¾d = savings on bill from PV sys. ¾t = 25 yrs ¾r = discount rate ¾5.3%/yr, OMB std.
¾re= growth rate for electricity cost ¾also 5.3%/yr
26
Why two discount rates? ¾ One (r) is used to calculate the nominal value of future costs/savings in year-zero dollars relative to an alternative investment ¾ The other (re) is used to model the increased nominal value of electric power over time (relative to all other goods). ¾ If r=re, then current value of the savings stream is the simple sum of yearly values. ¾ It’s difficult to predict either value! 27
Payback Time (PG&E) Plots vary by residential ‘demand factor’ ¾ Additional demand of the PHEV decreases PV payback time for customers with relatively low household electricity use ¾ The greater the customer’s base electricity use (demand factor), the more quickly the PV system pays for itself ¾ As a customer’s electricity use increases, the PHEV no longer significantly improves PV payback time
28
Payback Time (SCE) ¾ The effects of varying insolation and electricity demand do change the PV payback time ¾ These effects are minor compared to the dependence on rate structure ¾ SCE and SDG&E have non-tiered EV-TOU rate structures which negate the benefit of a PHEV on PV payback time ¾ PG&E’s rate structure increases the value of PV when a PHEV is included
29
Effect of PHEV + PV on CO2 ¾ Same 1 million PHEV and PV system scenario. ¾ 0.97 lbs of CO2 per kWh of grid charging from weighted average of CA electricity generation. ¾ 23.8 lbs of CO2 per gallon combusted gasoline. ¾ Car can drive 4 miles per 1 kWh of battery power. ¾ 20 mi of electricity only driving per day per car. ¾ Replaced car with 25 mpg fuel efficiency. ¾ Found that PHEV saves approximately 0.7 lbs of CO2 per electricity only mile driven. ¾ Scenario would save 0.003ppm per year of CO2 from entering the atmosphere.
Source Avoided Gasoline Driving PHEV Charging PV Systems Net Change
CO2 (lbs) CO2 (ppm) -6.95E+09 1.44E+09 -8.30E+09 -1.38E+10
-0.0015 3.10E-04 -0.0018 -0.003
30
