Jeff Smith, Matt RylanderEPRI

IEEE PES 2014, Washington, D.C.7/29/2014

PV Hosting Capacity on Distribution Feeders

2

Key Factors that Determine How Much PV a Feeder Can Accommodate (Hosting Capacity)

Large Scale PV Near Sub

Large Scale PV @ End Line

Size of PV Location of PV Feeder characteristics Electrical proximity to

other PV Unique solar resource

characteristics in the area

PV control 105%100%

95%

Distance from Substation

Vol

tage

VoltageRisefrom PV

Headroom

3

Key Requirements to Determine Hosting Capacity Distribution characteristics

Voltage response Short-circuit response Locational information Accurate feeder Models Proxy feeders arent sufficient

PV characteristics Location Size What about time?

Feeder Voltage Heat

Map

*all simulations performed with OpenDSS

4

Feeder Hosting Capacity: A Brief Primer

Baseline No PV

PV Penetration 1

PV Penetration 2

PV Penetration 3

Beyond

Increase Penetration Levels Until Violations

Occur voltage protection power quality thermal

PV Systems

Process is repeated

100s of times to capture

many possible

scenarios

5

Hosting CapacityOvervoltage Results Shown for Feeder J1

2500 cases shownEach point = highest primary voltage

ANSI voltage limit

Max

imum

Fee

der V

olta

ges

(pu)

Increasing penetration (kW)

Minimum Hosting CapacityMaximum Hosting Capacity

Total PV: 540 kW

Total PV: 1173 kW Voltage violation

No observable violations regardless of size/location

Possible violations based upon size/location

Observable violations occur regardless of size/location

6

Industry-Wide Assessment

Collaborative Effort

>30 feeders to date throughout US

Methodology Consistent Transparent

Open-source tools Advanced application of

traditional planning techniques

Details on analysis method: Stochastic Analysis to Determine Feeder Hosting Capacity for Distributed Solar PV. EPRI, Palo Alto, CA: 2012. 1026640.

7

Sample Results from Single feederSmall-Scale (Residential/Commercial)

Characteristic ValuekV 12PkLd 6.2MinLd 0.62TotalRegs 1Setpoint 1.0Band 4.0TotalCaps 1Totalkvar 1200EndofLineZ 15.88AvgZ 5.86MinZ 1.11MaxXR 7.87AvgXR 2.52MinXR 0.70TotalMiles 71.87TotalCustCount 1140.00EndofLineLength(mi) 11.07AvgR 2.16EndofLineMVA 9.10MinHeadrom 0.03LoadCenterR 5.90

Feeder Characteristics Hosting Capacity Results

8

Summary of Results from 18 Feeders

Research results found here:Distributed Photovoltaic Feeder Analysis: Preliminary

Findings from Hosting Capacity Analysis of 18 Distribution Feeders. EPRI, Palo Alto, CA: 2013. 3002001245.

All penetrations in this region are acceptable, regardless of location

Some penetrations in this region are

acceptable, site specific

No penetrations in this region are acceptable, regardless of location

0 2 4 6 8 10

J1

R1

R2

R3

R4

T1

T2

G1

G2

G3

P1

P2

P3

P4

P5

D1

D2

D3

MWofConsumerPV

Feed

er

0 2 4 6 8

P3

P5

PVPenetration(MW)

Feed

er Location matters

Noissues Mayrequireupgrades Requiresupgrades

2 similar feeders

0 2 4 6 8

P3

P5

PVPenetration(MW)

Feed

er Location matters

Noissues Mayrequireupgrades Requiresupgrades

2 similar feeders

*voltage-based hosting capacity shown

9

Summary of Results from 18 Feeders

Research results found here:Distributed Photovoltaic Feeder Analysis: Preliminary

Findings from Hosting Capacity Analysis of 18 Distribution Feeders. EPRI, Palo Alto, CA: 2013. 3002001245.

All penetrations in this region are acceptable, regardless of location

Some penetrations in this region are

acceptable, site specific

No penetrations in this region are acceptable, regardless of location

0 2 4 6 8 10

J1

R1

R2

R3

R4

T1

T2

G1

G2

G3

P1

P2

P3

P4

P5

D1

D2

D3

MWofConsumerPV

Feed

er

0 2 4 6 8 10

J1

R1

R2

R3

R4

T1

T2

G1

G2

G3

P1

P2

P3

P4

P5

D1

D2

D3

MWofLargePV

Feed

er

Small-Scale Distributed Large-Scale Distributed

*voltage-based hosting capacity shown

10

Detailed Hosting Capacity Analysis Can load be used to predict hosting capacity?

0

0.5

1

1.5

2

2.5

3

3.5

0 2 4 6 8 10 12 14 16 18

Minim

umHostin

gCapacity(M

W)

PeakLoad(MW)

Not without knowledge of other feeder characteristics

No correlation between hosting capacity and peak load

11

0

1

2

3

4

5

6

J1 R1 R2 R3 R4 T1 T2 G1 G2 G3 P1 P2 P3 P4 P5 D1 D2 D3

MW

Feeder

MinimumHostingCapacity100%ofDaytimeMinimumLoad

How Effective are Current Screening Practices? Keeping in mind, screens should be conservative by nature

0

0.5

1

1.5

2

2.5

3

3.5

J1 R1 R2 R3 R4 T1 T2 G1 G2 G3 P1 P2 P3 P4 P5 D1 D2 D3

MW

Feeder

MinimumHostingCapacity15%ofMaximumLoad

15% peak and 100% minimum load overestimates hosting capacity on many feeders

Feeders at risk for PV to pass through screens without issues being flagged

12

Correlation to Feeder CharacteristicsSmall-Scale PV, Primary Voltage Deviation Hosting Capacity

Strong correlation Number of line

regulators Total circuit length End of line length Load Center

Location PV Center

Location

13

Next Step Applications

Developing better screening techniques for solar PV

Developing streamlined methods for determining hosting capacity Application across entire

service territory (100s to 1000s of feeders)

Commercial software Identifying technical solutions for

accommodating higher penetration levels Streamlined

methods

14

Conclusions

Consistent and repeatable methodology allows greater understanding of impacts due to Location of PV Feeder characteristics

Many of issues start arising at secondary level Models typically dont go there!

Trends in hosting capacity dependency are found Regulation, line impedance, and load location Load level not so much

Other factors play in as well How close to thresholds in basecase (headroom)

15

Questions

Contact:

Jeff SmithManager, Power System StudiesEPRIjsmith@epri.com