PREDICTABLE PROFITABILITY

© 2019 OTT HydroMet 1© 2019 OTT HydroMet 1

Presenters: Damon Nitzel, Troy Morlan, Carla Dawson

June 16th, 2021, Hosted by PV Magazine

PREDICTABLE PROFITABILITY:THE IMPORTANCE OF SOILING MONITORING & MITIGATION

© 2019 OTT HydroMet 2© 2019 OTT HydroMet 2Meteorology Division

AGENDA

Panel soiling variability

Monitoring technologies

Hardware & software integration

Cleaning methods

100 MW case study

Suggested methodology

© 2019 OTT HydroMet 33Meteorology Division

SITE PRODUCTION IS NOT MEETING EXPECTATIONS

Solar projects in 2016-2019 underperformed their target production (P50 estimates) on

average by 6.3%.

Richard Matsui. Solar Generation Index: Quantitative Insights from Leading Asset Owners. 2020. kWh Analytics

PV site analyses are finding that module soiling is a leading cause of recoverable

underperformance.


SURVEY:

IN YOUR EXPERIENCE, WHAT % OF ANNUAL REVENUE COULD

YOU RECOVER FROM SOILING COSTS?

A. Unrecoverable (0%)

B. Less than 2%

C.2 to 4%

D.4 to 7%

E. More than 7%


PV PANEL SOILING

A definition:

Surface accumulation of material(s) that block or

scatter incident light, decreasing PV module power

output.

Predicting the cost of being dirty

Soiling rate/monitoring

PPA price

Solar availability (irradiance)

Cleaning resources

Energy lost

A 100 MW case study

Estimated lifetime loss: 3.2 million dollars (1.3%)


SOILING VARIATION

Left image and regional soiling values modified from:Klemens Ilse, Leonardo Micheli, Benjamin W. Figgis, Katja Lange, David Daßler, Hamed Hanifi, Fabian Wolfertstetter, Volker Naumann, Christian Hagendorf, Ralph Gottschalg, Jörg Bagdahn,Techno-Economic Assessment of Soiling Losses and Mitigation Strategies for Solar Power Generation , Joule, Volume 3, Issue 10, 2019, Pages 2303-2321, ISSN 2542-4351https://doi.org/10.1016/j.joule.2019.08.019.(https://www.sciencedirect.com/science/article/pii/S2542435119304222)

Note: Soiling rates are in average percent (%) per day.

Case study data, Almeria, Spain

0.2% 0.2%

0.15%

0.4%

1.3%

0.38%

0.03%

0.5%

0.35%0.15%

0.15%

0.4%

0.08%


INTRA-SITE VARIABILITY EXAMPLE

https://mesonet.agron.iastate.edu/sites/windrose.phtml?station=WJF&network=CA_ASOS

Wind Rose shows typical wind

blows from WSW

Estimated 30 MW solar farm east

of agricultural field

Wind(WSW)

Higher soiling areas

N

https://mesonet.agron.iastate.edu/sites/windrose.phtml?station=WJF&network=CA_ASOS


SOILING RATIO DEFINITION

Soiling Ratio: 𝑆𝑅 =𝐼𝑆𝐶,𝑆

𝐼𝑆𝐶,𝐶

(𝑆𝑜𝑖𝑙𝑒𝑑)

(𝐶𝑙𝑒𝑎𝑛)

ISC,S short circuit of soiled panel

ISC,C short circuit of clean panel

Transmission loss: 𝑇𝐿 = 1 − 𝑆𝑅


SUMMARY OF MEASUREMENT TECHNOLOGIES

Two-panel clean & dirty system Optical sensing


INTERNAL SIGNAL OF REFLECTED LIGHT


OPTICAL SENSING INNOVATIONS

Multi-sensor analysis

No routine maintenance

Cleaned the same as the PV array

No moving parts or daily/weekly water

Representative sampling surface

Common module glass and frame size

Mounted in-array with panels

Soiling drifting (via rain, dew, snow)

Similar wind flow (deposition/clearing)


HOW IS SOILING MEASURED?

Dust IQ soiling station is installed in the field onto solar panel array

Soiling station is connected via communication cabling to MET station communications enclosure


HOW IS THE DATA COLLECTED?

Communication protocols between soiling station and cloud server

MODBUS RTU MODBUS TCP OPC UA

Dust IQ Sensor in the fieldMET station enclosure in the field

SCADA rack in the field/substation

Cloud-hosted HMI/historian server


MET STATION AND SCADA INTEGRATION

HMI screenshot for soiling levels


WHAT WAYS CAN THIS DATA BE USED?

Custom alarms can be configured

into the overall SCADA system to

show when soiling levels reach

impactful thresholds (determined

by soiling percentages)

O&M teams can determine the

set points for each site or array

where the energy generation loss

from soiling levels is greater than

cleaning costs


WHEN DO YOU CLEAN? (POLL)

A. Lost performance site visit, observe dirty panels

B. Set schedule (e.g. 1x, 2x per year)

C.Use soiling data to calculate maximum ROI

D.Never


SOLAR PANEL CLEANING

AUTOMATION AND ROBOTICS

Manual Semi-automatic

XY Semi. Tractor Semi. X-only Semi.

Automatic

Auto. & Smart

Operator driven

Manpowered


WHEN DOES FULL AUTOMATION MAKE SENSE?

Examples of detailed engineering study considerations

Higher soiling rates

Personnel safety

More remote location

Low water resource

Longer rows

Panel sizes

# of panels stacked

…


PANEL CLEANING SEMI-AUTOMATION VIDEO


CASE STUDY: INTRODUCTION

Size: 100 MW

PPA Value: $28 MWh

Monitoring Type: Optical sensor (DustIQ)

Cleaning Type: Manual compared to Semi-automatic

Yearly soiling loss: 2.86% (yearly average)

Cleaning Frequency: 1 per year

Conclusion:

Monitoring + Automation could have reduced the cost of being dirty by 1.3%.

Solar PV site lifetime (30-year) impact: $3,600,000 margin increase


Total cost of soiling & cleaning KPI:

No action (est.): $272,400

1 Manual wet brush: $362,600

Result: 1 cleanings/year

“Savings”: -$90.2K/year, -$2.7M over lifetime (30-yr)

CASE STUDY: WET BRUSH REVENUE LOSS

Manual Cleaning

Natural (rain) Cleaning

*PPA value: $28/MWh (CAISO), Site Size:100 MW

$-

$50,000.00

$100,000.00

$150,000.00

$200,000.00

$250,000.00

$300,000.00

$-

$500.00

$1,000.00

$1,500.00

$2,000.00

$2,500.00

$3,000.00

$3,500.00

$4,000.00

$4,500.00

$5,000.00

0 100 200 300 400

Cu

mu

lati

ve R

even

ue

[U

SD]

Rev

enu

e Lo

ss/D

ay [

USD

]

Time [Days]


No Action

Manual Semi-automated

Wet Brush Linear Robot Tractor

Cleaning frequency (per year) 0 1/yr 3/yr 3/yr

Cleaning CapEx $ - $ -$ 160,000

$ 315,000

Cleaning OpEx $ -$ 2,000,000

$ 657,000

$ 185,000

Total Cleaning Cost $ -$ 2,000,000

$ 817,000

$ 500,000

Module soiling loss$ 2,723,000

$ 1,626,000

$ 1,067,000

$ 998,000

Total soiling & cleaning cost$ 2,723,000

$ 3,626,000

$ 1,884,000

$ 1,497,000

Savings (compared to no action) $ - $ (904,000) $ 840,000 $ 1,226,000

Total revenue (zero soiling)$ 95,046,000

$ 95,046,000

$ 95,046,000

$ 95,046,000

Total soiling & cleaning cost (% of rev.) 2.9% 3.8% 2.0% 1.6%

CASE STUDY: 10-YEAR FINANCIAL ESTIMATES

Average soiling rate est.: 0.08%/dayAverage yearly soiling: 2.86%


No Action





$ 315,000


$ 657,000

$ 185,000


$ 817,000

$ 500,000


$ 1,626,000

$ 1,067,000

$ 998,000


$ 3,626,000

$ 1,884,000

$ 1,497,000



$ 95,046,000

$ 95,046,000

$ 95,046,000





SUGGESTED METHODOLOGY

Financial Modeling

Engineering & Development

Construction & Commissioning

Operations & Maintenance

Performance Analysis

Measure soiling

during site

assessment

Action:

Effect: Decrease

underperformance

risk



Financial Modeling





Design for

highest ROI

cleaning

method

Measure soiling

during site

assessment

Action:

Effect: Decrease

underperformance

risk

Decrease total

site costs by

1.3% ($3.2M)*

*Based on case study presented, 30-year site life



Financial Modeling





Integrate soiling

monitoring into

met station &

ops center

Design for

highest ROI

cleaning

method

Measure soiling

during site

assessment

Action:

Effect: Decrease

underperformance

risk

Decrease total

site costs by

1.3% ($3.2M)*

Data available

to inform

decisions




Financial Modeling





Use real-time

data to decide

when to clean

Integrate soiling

monitoring into

met station &

ops center

Design for

highest ROI

cleaning

method

Measure soiling

during site

assessment

Action:

Effect: Decrease

underperformance

risk

Decrease total

site costs by

1.3% ($3.2M)*

Data available

to inform

decisions

Higher

profitability

year-to-year




Financial Modeling





Use real-time

data to decide

when to clean

Integrate soiling

monitoring into

met station &

ops center

Design for

highest ROI

cleaning

method

Measure soiling

during site

assessment

Report savings

& accurate

soiling losses

Action:

Effect: Decrease

underperformance

risk

Decrease total

site costs by

1.3% ($3.2M)*

Data available

to inform

decisions

Higher

profitability

year-to-year

Inform future

production

models



Total cost of soiling & cleaning KPI:

No action (est.): $272,400 (2.9%)

1 Manual wet brush: $362,600 (3.8%)

3 Tractor cleanings: $149,800 (1.6%)

Designed Result: 3 cleanings/year

Savings: $122.6K/year, $3.6M over lifetime (30-yr)

CASE STUDY: DESIGNED FOR TRACTOR CLEANING

Tractor Cleaning

Natural (rain) Cleaning5%

soiled

*PPA value: $28/MWh (CAISO), Site Size:100 MW

$-

$50,000.00

$100,000.00

$150,000.00

$200,000.00

$250,000.00

$300,000.00

$-

$500.00

$1,000.00

$1,500.00

$2,000.00

$2,500.00

$3,000.00

$3,500.00

$4,000.00

$4,500.00

$5,000.00

0 100 200 300 400

Cu

mu

lati

ve R

even

ue

[U

SD]

Rev

enu

e Lo

ss/D

ay [

USD

]

Time [Days]


USING ON-SITE SOILING DATA, WHICH CLEANING

METHOD WOULD YOU USE?

A.I don’t know, I haven’t determined it yet

B.No action (natural)

C.Manual Wet Brush

D.Semi-Automated On Array Robot

E.Semi-Automated Tractor


WHAT WE’VE LEARNED TODAY

Case study KPI review:

No action: 2.9% (revenue loss per year); $2.7M/10-yr

Manual: 3.8%; $3.6M/10-yr

Linear Robot: 2.0%; $1.9M/10-yr

Tractor: 1.6%; $1.5M/10-yr

High soiling variability requires local, site-specific

Monitoring soiling reduces financial risk

Designing for optimal cleaning increases site profitability


WE’RE AVAILABLE TO HELP

Damon NitzelApplication Development Manager, Solar

Email: [email protected]

Troy MorlanLead Eng. Proposal Coordinator


Carla DawsonChief Innovation & Growth Officer


PREDICTABLE PROFITABILITY

Documents

Transcript of PREDICTABLE PROFITABILITY