Distribution Costs and Benefits of Increasing Penetrations of PV
Jeff Smith, EPRI,
[email protected] Nadav Enbar, EPRI,
[email protected] USC 2015 San Diego, April 27 - 29
© 2015 Electric Power Research Institute, Inc. All rights reserved.
Outline Distribution systems are unique Summary of cost/benefit results from a range of feeders Evaluating distribution “system-wide” impacts challenges and solutions
2 © 2015 Electric Power Research Institute, Inc. All rights reserved.
Distribution Systems are Unique and Large in Scale
– Designed to reliably serve ALL customers in least-cost manner – Evolve around customer needs – Feeder design impacts reliability, costs, and benefits of DER
Ex: PV Hosting Comparison by Feeder Feeder
Distribution Feeders are Unique
D3 D2 D1 P5 P4 P3 P2 P1
Distribution Systems are Vast
G3 G2
G1 Typical Distribution Utility
Number
Service Territory
1
Planning area
1’s - 10’s
Substations
10’s - 100’s
Feeders
100’s -1000’s
Service Transformers
1000s 1,000,000’s
Customers
100,000’s 1,000,000’s
T2 T1
Decreasing visibility
R4 R3 R2 R1 J1 0
1
2
3
4
5
MW of Consumer PV
3 © 2015 Electric Power Research Institute, Inc. All rights reserved.
Integrated Grid Methodology Distribution System
To fully realize the impact and value of DER a deliberate and beneficial integration is needed Widespread deployment of PV needs to be incorporated into both grid planning and operational processes
Hosting
Energy
Capacity
Reliability Benefit/Cost
Scenario Definition Market Conditions
DER Adoption
System Assumptions
EPRI has developed a comprehensive benefit-cost framework aimed at better informing strategic decisionmaking
4 © 2015 Electric Power Research Institute, Inc. All rights reserved.
Bulk System Resource Adequacy
Transmission Performance
Flexibility
Transmission Expansion
Operational Practices & Simulation
An Integrated Grid
System Benefits
Societal Costs/Benefits
System Costs
Customer or Owner Cost/Benefits
Method for Distribution Analysis Determines the true costs and benefits of DER to the distribution system.
voltage
Voltage
1st step: PV and Feeder Performance
Protection relay desensitization
Load Only
Current
Watts
limit
time
Capacity
– Hosting capacity (voltage and protection)
unacceptable overvoltage
Load and PV
– Thermal capacity
Impedance
Integrated Approach
Impedance
– Energy (losses and consumption) – determine the value (costs and benefits) of the resource to the grid.
Energy
2nd Step: Cost/benefit
Energy Losses
unserved energy
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Month
Energy
5 © 2015 Electric Power Research Institute, Inc. All rights reserved.
energy exceeding normal
Time
Reliability
Example Case Studies
Cost/Benefit Results from Three Feeders
6 © 2015 Electric Power Research Institute, Inc. All rights reserved.
Feeder Results from Hosting Capacity Analysis Feeder K3
Feeder S1
Feeder Evaluation Criteria
Feeder K2
3 distribution feeders selected based on their range in characteristics and hosting capacities.
Protection & Voltage All penetrations in this region are acceptable, regardless of location
Protection
Some penetrations in this region are acceptable, site specific
Voltage
No penetrations in this region are acceptable, regardless of location
7 © 2015 Electric Power Research Institute, Inc. All rights reserved.
Thermal Capacity Analysis What is considered? Watts
– Capacity deferral: analyzed based on utility-specific practices for planning system upgrades.
Load and PV
– Thermal aging: analyzed based on IEEE Standard 57.91 for thermal aging calculations.
– Historical load and solar data. – Transformer thermal characteristics and ratings. – Ambient conditions.
Time 0.025
Cummulative Loss of Life (%)
Quantification of impact based on:
Load Only
0.020
Increasing levels of PV
0.015
0.010
0.005
0.000 0
1
2
3
4
Year (2% load growth/year)
*example illustrations only
8 © 2015 Electric Power Research Institute, Inc. All rights reserved.
5
6
Application of Thermal Capacity Assessment Method to an Example Feeder 8000
K2 7000
Normal Planning Limit of 6,600 kVA
Available capacity
Power (kVA)
6000 5000 4000 3000
•
Yearly load profile analyzed, peak load day/level identified for each
•
Planned load growth applied
2000
1000 0
Jan 3
Time (Hours)
Jan 4
Jan 5 Data
Feeder K2
Current Loading (% of normal limit)
82%
Planned Load Growth Calculated Load Growth Until Capacity Constrained 9 © 2015 Electric Power Research Institute, Inc. All rights reserved.
1%/year ~20 years
Application of Thermal Capacity Assessment Method Continued Statistical Distribution of Hourly Output - K2 100%
Normalized PV Power (% of PV Rating)
90%
K2 Feeder
peak load hour
80% 70%
Max and Min Median Inner quartile (25th and 75th)
60% 50% 40% 30% 20% 10% 0% 5
6
7
8
9
10
11
12
13
14
Hour of Day
15
16
17
18
19
20
21
Dec 1-Feb 29, 2011
In order to defer capacity, the resource (PV) must be available when the asset is most constrained
10 © 2015 Electric Power Research Institute, Inc. All rights reserved.
Thermal Capacity Assessment: Summary of Results for Three Feeders The differences in deferral potential were driven primarily by the available capacity on the feeder and the utility planning practices employed. Technical Metrics Current Loading (% of normal limit) Planned Load Growth Calculated Load Growth until Capacity Constrained
Feeder K2
Feeder K3
Feeder S1
82%
67%
64%
1%/year
0.5%/year
1%/year
~ 20 years
~ 80 years
~ 50 years
$13,050
0
0
Economic Metrics Deferral Savings
11 © 2015 Electric Power Research Institute, Inc. All rights reserved.
Energy Analysis Line and core losses
What is considered?
– Energy loss and consumption on feeder with no PV. – Energy loss and consumption on feeder with PV.
Quantification of impact based on: – Difference in losses and consumption with and without PV. – Annual simulation using measured load profiles and local ground-based PV measurements to drive simulations. 12 © 2015 Electric Power Research Institute, Inc. All rights reserved.
Low-voltage line losses
Customer meter
Analysis Metric
Transmission
– Change in losses on distribution feeder resulting from PV. – Change in consumer energy consumption due to changes in delivery voltages.
Medium voltage line losses
Transformer load and noload losses
Energy consumption
Energy Assessment: Summary of Results Continued Line losses are more greatly impacted (reduced) for feeders with greater initial losses Change in consumption can impact net losses At higher penetration levels, improvement in losses are lessened – Increased consumption on feeder – Reverse power flow
Feeder K2 4.10% 3.14
Feeder K3 2.70% 5.3
Feeder S1 6.30% 4.6
0.70%
0.32%
0.71%
New Losses with PV
3.40%
2.38%
5.59%
% Change in losses/MW of PV
-5.44%
-2.24%
-2.45%
Net Reduction in Losses with PV (line & core & consumption)
0.47%
0.07%
0.10%
New Losses with PV (Net) % change in Net Losses/MW of PV
3.63% -3.65%
2.63% -0.49%
6.20% -0.35%
Base Losses without PV PV Capacity (MW) Reduction in Losses with PV (line & core)
13 © 2015 Electric Power Research Institute, Inc. All rights reserved.
Mitigation Analysis (Voltage and Protection) Impacts Considered
Voltage Mitigation Options Considered
– Voltage
• Reconductoring
– Protection
• Service transformer replacement • Service upgrade
Mitigation Measures – Prevent adverse voltage or additional control operations – Prevent inadvertent protection operation
Quantification of Mitigation
• Add voltage regulator • Smart inverters
Protection Mitigation Options Considered Directional relay/settings Reconductoring
– Existing utility upgrade practices
Grounding recloser/transformer
– Cost of mitigation
Direct transfer trip
Breaker replacement
14 © 2015 Electric Power Research Institute, Inc. All rights reserved.
Application of Voltage & Protection Assessment Method to Feeder K2 Three PV penetration levels (0.5/1.0/2.0 MW) Mitigation is needed at all three PV penetrations selected for consideration. More optimal PV deployments typically require less upgrades. 0.5 MW scenario 1.0 MW scenario 2.0 MW scenario Secondary Overvoltage Secondary Voltage Deviations
No Issues Primary Imbalance
May require upgrades
Regulator Voltage Change
Requires upgrades
Primary Voltage Deviations Primary Overvoltage
0 15
2 Hosting Capacity (MW)
4
© 2015 Electric Power Research Institute, Inc. All rights reserved.
*protection upgrades also needed at 2MW level
Application of Voltage & Protection Assessment Method Continued
Cost of voltage and protection mitigation options needed based on utility estimates for each Mitigation Means
Materials Installation
Units
Lifetime
3Ø Reconductor to 795AA
$ 50,000
$ 250,000
per mile
40
1Ø Reconductor to 795AA
$ 16,667
$ 125,000
per mile
40
3Ø Regulator
$ 40,000
$ 35,000
each
30
Capacitor Bank
$10/kvar
$ 1,000
each
30
Distribution Transformer
$ 1,000
$ 600
each per Ø
40
Line Recloser
$ 30,000
$ 20,000
feet
40
Secondary/Service Upgrade
$ 60 + $2/ft
$ 80+$3.2/ft
feet
40
Assumed in 2014$ 16 © 2015 Electric Power Research Institute, Inc. All rights reserved.
O&M
$2000/yr
Summary of Example Feeder Results for Three Feeders Feeder K2 Capacity Deferral 0.5 MW 1 MW 2 MW
-.15 ȼ/kWh -.15 ȼ/kWh -.15 ȼ/kWh
Feeder K3
Feeder S1
0
0
Mitigation Costs 0.5 MW 1 MW 2 MW
.01 to 2 ȼ/kWh .01 to 5 ȼ/kWh 1 to 7 ȼ/kWh
Losses (line & core)
5.4% per MWPV
2.2% per MWPV
2.4% per MWPV
Losses (net)
3.6% per MWPV
0.5% per MWPV
0.3% per MWPV
0
0.64 ȼ/kWh 0.3 to 0.8 ȼ/kWh 0.2 to 0.8 ȼ/kWh
Each feeder has a unique benefit and cost based on its particular characteristics and utility planning methods.
17 © 2015 Electric Power Research Institute, Inc. All rights reserved.
Assessing DER Across Entire Distribution System Challenge – Distribution feeder impacts are unique
Substation Level
=
– Distribution system is vast – Issues/value streams we want to identify are unique and at the “edge” of the grid – Need to consider unique response at the individual feeder level
EPRI Approach – New method that captures impact and value efficiently w/o sacrificing accuracy – Individual feeder response level
Feeder Level
Feeder Level
– Can be applied throughout entire distribution system 18 © 2015 Electric Power Research Institute, Inc. All rights reserved.
Feeder Level
Assessing DER Across Entire Distribution System Challenge – Distribution feeder impacts are unique
System-Wide Assessment Capturing Feeder-Specific Results
– Distribution system is vast – Issues/value streams we want to identify are unique and at the “edge” of the grid – Need to consider unique response at the individual feeder level
EPRI Approach – New method that captures impact and value efficiently w/o sacrificing accuracy – Individual feeder response level – Can be applied throughout entire distribution system 19 © 2015 Electric Power Research Institute, Inc. All rights reserved.
Distribution-Wide Assessment of DER EPRI Streamlined Hosting Capacity Method
Utilizes current utility planning tools and data (no new tools are needed) Evaluates each feeder individually Can be applied throughout entire system (1000’s of feeders) in automated fashion Considers “feeder-level” response with results that can be aggregated up to substation level Captures value/cost efficiently w/o sacrificing accuracy Applications – Distribution resource plans – Screening – …. Distribution diagrams courtesy of Salt River Project 20 © 2015 Electric Power Research Institute, Inc. All rights reserved.
Conclusion Voltage
Thermal Capacity
New EPRI approach allows for system-wide assessment without sacrificing necessary details
limit
time
Protection relay desensitization
Current
Load Only
Distribution systems are unique in terms of the cost/benefits of DER
Load and PV
Impedance
Integrated Approach
Impedance
unserved energy
Energy Losses
Assessing DER impacts across entire system can be challenging
unacceptable overvoltage
Energy
Watts
Distribution systems are unique in terms of their response to DER
voltage
Must consider where DER connects – often at customer level
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Month
Energy
21 © 2015 Electric Power Research Institute, Inc. All rights reserved.
energy exceeding normal
Time
Reliability
Questions Reference Understanding the Costs and Benefits of Increasing Grid Penetrations of Solar Photovoltaics. EPRI, Palo Alto, CA: 2014. 3002003270. Distributed Photovoltaic Feeder Analysis: Preliminary Findings from Hosting Capacity Analysis of 18 Distribution Feeders. EPRI, Palo Alto, CA: 2013. 3002001245. A New Method for Characterizing Distribution System Hosting Capacity for DER: A Streamlined Approach for PV. EPRI, Palo Alto, CA: 2014. 3002003278. The Integrated Grid: A Benefit-Cost Framework, EPRI, Palo Alto, CA: 2015. 3002004878 22 © 2015 Electric Power Research Institute, Inc. All rights reserved.