EPRI Q3 2014 Meeting - Batteries & Power Optimizers

1. Battery Integration with Solar Power Plants 2. Power Optimizer

Jake McKee, Head of engineering Solar PV Stefan Christ, Manager R&D Solar EPRI Fall Meeting, Atlanta, Sept 29th 2014

Agenda

Battery Integration with Solar Power Plants Jake McKee Power Optimizer test results Stefan Christ Q & A Stefan & Jake

2 EPRI Fall Meeting, Battery integration and Power Optimizer, Sep 29th, 2014

Introduction to E.ON’s PV business

E.ON‘s PV Business:

~80 MW of utility scale installations in Europe and in the U.S. plus 40 MW currently under construction

Rooftop PV business in Europe

E.ON‘s Objective is to make PV more attractive

E.ON constantly strives to cut LCOE for its PV installations

Different options like optimizing tracking as well as other advance concepts Power Optimizer (POs) explored to cut LCOE for utility scale and to

increase flexibility in residential systems Integrating batteries and PV to improve the value of PV power to energy

system, e.g. better control (reducing ramp rates), ancillary services.

3 EPRI Fall Meeting, Battery integration and Power Optimizer, Sep 29th, 2014

Applications & Opportunities

Optimization of Solar Projects without Batteries

Experiences with Battery Systems and Requirements of Puerto Rico

Optimizing Solar Projects including Battery Energy Storage Systems (BESS)

Battery Integration with Solar Power Plants – Overview

EPRI Fall Meeting, Battery integration and Power Optimizer, Sep 29th, 2014 4

APPLICATIONS AND OPPORTUNITIES BATTERIES (and solar projects)

EPRI Fall Meeting, Battery integration and Power Optimizer, Sep 29th, 2014 5

6 EPRI Fall Meeting, Battery integration and Power Optimizer, Sep 29th, 2014

OPTIMIZING - SOLAR PROJECTS

EPRI Fall Meeting, Battery integration and Power Optimizer, Sep 29th, 2014 7

LCOE/IRR/NPV Optimization of Solar PV Plants

We optimize for IRR, LCOE, NPV

What are we optimizing?

Technology

FT, SAT, 2-Axis

Mono, Poly, Thin Film, CPV

Ground Coverage Ratio (GCR)

Inverter Load Ratio (ILR)

AC Size of the Plant

Procurement support Similar tools, slightly different concept

EPRI Fall Meeting, Battery integration and Power Optimizer, Sep 29th, 2014 8

But.. It is not an Exact Science!

Complications:

Accuracy of modeling efforts

Accuracy of the meteo file

How good are CAPEX estimates for years in the future?

Quality of OPEX estimates EPRI Fall Meeting, Battery integration and Power Optimizer, Sep 29th, 2014 9

PUERTO RICO EXAMPLE

Required Combining Batteries with Solar Projects

EPRI Fall Meeting, Battery integration and Power Optimizer, Sep 29th, 2014 10

PREPA (Puerto Rican Electric Power Authority) and MTRs!!

Ramp Rate + Frequency Control

What if these happen at the same time?

MTRs led to a cost benefit sizing of battery

Complex language to measure violations

How many violations do we accept

EPRI Fall Meeting, Battery integration and Power Optimizer, Sep 29th, 2014 11

Ramp Rate Control

The PV facility shall be able to control the rate of change of power output

Rate of decrease of power!

A 10 % per minute rate (based on AC capacity) limitation enforced

Is independent of meteorological conditions

EPRI Fall Meeting, Battery integration and Power Optimizer, Sep 29th, 2014 12

Frequency Response

The PV facility shall provide an immediate real power primary frequency

response of at least 10% of the maximum AC active power capacity

The time response (full 10% frequency response) shall be less than 1 second

The facility frequency response shall be maintained for at least 9 minutes

13 EPRI Fall Meeting, Battery integration and Power Optimizer, Sep 29th, 2014

OPTIMIZING – SOLAR PROJECTS AND BATTERIES

EPRI Fall Meeting, Battery integration and Power Optimizer, Sep 29th, 2014 14

Selecting the BESS Technology for your Project

Lead Acid Low Efficiency Low Cost 100% Recyclable

Flow technologies Benefit from higher cycle lifetimes Over 4 hours do flow battery Break point to go to flow is 2 hours

NaS High energy to capacity ratios allow performance as a

capacity resource, as backup generation to avoid outages at remote sites, and to defer transmission and distribution (T&D) asset upgrades on overstressed lines and substations Also capable of performing frequency regulation services

NaS and Flow Worse roundtrip efficiency than Li-Ion technologies

Li-Ion High energy density Microsecond response time EPRI Fall Meeting, Battery integration and Power Optimizer, Sep 29th, 2014 15

Contracting the BESS (Integrator and Battery Decisions)

Can the battery supplier do the integration?

If they work with an integrator, is there a wrap guarantee?

Developer prefers one contract; with a large balance sheet

What are their maintenance plans?

What if BESS degrades faster than the solar field and it’s

requirements?

If tied to a 20+ year solar PPA Battery replacement plan

A not to exceed price

Leads to an all inclusive NPV evaluation of suppliers in the RFP

EPRI Fall Meeting, Battery integration and Power Optimizer, Sep 29th, 2014 16

Financial Model – Accounting for Your Combined Solar and Battery

The “round-trip efficiency” of the

battery system –

These are losses from the solar

production

Should be accounted for in

financial model

While minor, batteries also

dissipate when storing over periods

of time

Ancillary services contract

Or take it from the grid

Sign a utility contract

Statistical efficiency

through operating

projects

Guaranteed

efficiency

Battery (round-trip)

Inverter (round-trip)

Transformer (round-trip)

*only if vendor provide the

transformer.

Parasitic Load (round-trip)

EPRI Fall Meeting, Battery integration and Power Optimizer, Sep 29th, 2014 17

Agenda

Battery Integration with Solar Power Plants Jake McKee Power Optimizer test results Stefan Christ Q & A Stefan & Jake

18 EPRI Fall Meeting, Battery integration and Power Optimizer, Sep 29th, 2014

Three tests and three goals form the core of E.ON’s PO testing strategy

A) Basic understanding of PO technology and suppliers’ products B) Assessment of performance in shading conditions C) Assessment of impact on BoS savings for large scale application

19

Retro-fit Rooftop Tests Midsize Plant

“Retro-fit” Ground farm Test Utility Scale

New build Rooftop Test Residential Scale

A) B)

B) C) 3.

1.

2.

EPRI Fall Meeting, Battery integration and Power Optimizer, Sep 29th, 2014

Midsize Retrofit Field Tests - Boundary Conditions

Existing commercial roof-top PV plant

99 kWp fixed structure

9 subfields each with 3 strings á 16 modules on 10 kVA inverter plus one 2 kVA inverter

Inverter minimum DC voltage: 333 V

20

POs from three different suppliers tested

2x „Buck & Boost“, 1x „Buck“ type

For each PO supplier one full subfield equipped

6 non-retrofitted subfields available for reference evaluation

1.

EPRI Fall Meeting, Battery integration and Power Optimizer, Sep 29th, 2014

Mixed performance results in regular operation

21

Pre-/Post-Retrofit Performance Comparison* vs. Reference Subfields

Overall retrofitted subfields show similar performance compared to “normal” subfields

Disadvantage for retrofitted subfields during low yield generation

Advantage for retrofitted subfields during high yield generation

* refers to daytime period 10 a.m. to 4 p.m. ** share of total yield: “<5 kWh”-hours approx. 36-38%, “>5 kWh”-hours approx. 62-64%, *** valid comparison data for end of 2014 envisaged

PO 1(Buck & Boost type)

PO 2(Buck type)

PO 3(Buck & Boost type)

Total 1.0% 0.0%<= 5 kWh** -3.0% -6.0%

> 5 kWh** 4.0% 4.0%

Pre-/Post-PO-Retrofit Comparison (vs. Reference)

no valid comparison data available ***

Criteria:Hourly Yield*

1.

EPRI Fall Meeting, Battery integration and Power Optimizer, Sep 29th, 2014

Only sunny days used for near shading performance testing

“Extreme” shading conditions, i.e. no diffuse irradiance on shaded cells

Several shading scenarios relevant for utility scale plants simulated, e.g. A) Row-to-Row shading for landscape module orientation C) Several single shading objects in each string

PO advantage vs. shaded reference ranging 0-35% for Buck & Boost 0-21% for Buck only

22

C)

PO 1(Buck & Boost type)

PO 2(Buck type)

PO 3**(Buck & Boost type)

Shaded Reference

A 77.0% 41.5% 76.5% 42.0%C 76.5% 71.5% 74.5% 50.5%

ShadingScenario

Subfield Yield vs. non-shaded Reference*

* Shading tests results refer to a single day for each shading scenario; Comparison normalised, refers to daytime period 10 a.m. to 4 p.m. ** only very limited pre-retrofit data available

Good performance results in near shading conditions 1.

A)

EPRI Fall Meeting, Battery integration and Power Optimizer, Sep 29th, 2014

Comparative tests being performed on East-West-rooftop in Germany (1 out of 5 test sites)

modules with PO

modules without PO

PV System A PV System B POs: 24 DC/DC Optimiser no PO Module Type: 240 Wp mono Si Modules allocation: each 12 East / 12 West PV capacity: each 5.8 kWp Inverter: each 3.7 kW, 2 MPPT

23

2.

EPRI Fall Meeting, Battery integration and Power Optimizer, Sep 29th, 2014

Performance impacted by a tree’s shadow in afternoon

25th March 2014

Power optimisers’ positive impact on output in near shading condition

West side observation 20th March at 4.20 p.m

24

2.

EPRI Fall Meeting, Battery integration and Power Optimizer, Sep 29th, 2014

Some time later the shade is gone and so is the DC/DC advantage

19th May 2014

Continuous slight underperformance of PO equipped system when no nearby shading occurs

25

2.

EPRI Fall Meeting, Battery integration and Power Optimizer, Sep 29th, 2014

Monthly performance does not show a Power Optimizer advantage

Monthly production

26

2.

EPRI Fall Meeting, Battery integration and Power Optimizer, Sep 29th, 2014

Summary of current PO testing

Results & Findings:

Test setups and provided no evidence for performance enhancement by POs in conditions with no near shadings

Performance enhancement by POs in near shading conditions observed

Next Steps:

Tests in large scale utility ground farm will focus on BoS optimization

Ongoing discussions with PO suppliers about further technology optimization

27

1. 2.

3.

EPRI Fall Meeting, Battery integration and Power Optimizer, Sep 29th, 2014

Thank You.. Questions?