Assessment of the IAV of the GHI in the Atacama Desert
of Chile
P Darez1,3, J Baudry1,2, C Darr1,3
1 Grupo Técnico de Energías Renovables2 Mainstream Renewable Power Chile3 350renewables SpA
Who is GTER?
• Renewable energy industry platform foundedby volunteers based in Chile
• Workstreams:• PV module soiling,
• IAV
• Wind turbine noise
• Reference station network
• (…)
Develop Best
Practice
Engage
stakeholders
Encourage
cooperation
Motivation
• Atacama: highest solar resource
• Several GW of PV in development
• IAV value significant for project
financing
• Current generic value (4-6%) may
be too conservative
Source: SolarGIS
Typical site in the Atacama?
Typical site in the Atacama?
Typical site in the Atacama?
Typical site in the Atacama?
Methodology
• 15x GHI measurements, up
to 3.7 years, 14 locations
• 14x SolarGIS® datasets,
13 to 14 years
• Validation of model with
measured data
• Calculation of IAV from
model data© OpenStreetMap contributors
SolarGIS® model validation
Site from North to South RMSE Daily [%] RMSE Monthly [%] Bias [%]
MRP 2 12.8 9.9 -5.5
MRP 1 5.6 1.9 -1.5
Pampa Camarones 4 1.5 -0.8
Pozo Almonte 5.3 3.8 1.8
MRP 3 4.2 2.5 -2.3
Crucero 3.3 1.3 -0.8
Salar 8.8 8.4 4.5
San Pedro de Atacama 4.6 2.7 2
Puerto Angamos 6.7 5 2.4
Cerro Armazones 4.6 2.8 -2.7
Salvador 2.7 1.5 -0.5
MRP 4 4.8 1.8 -1.4
Inca de Oro 3.5 1.9 -1.4
MRP 5 5.9 4.1 -3.5
MRP 6 5.5 1.6 -0.7
IAV calculation
Where:
- X = {Xyear 1, Xyear 2, Xyear 3, …} is the
considered variable, in this study monthly
or yearly mean GHI.
- s is the standard deviation of X
- E is the mean value of X.
IAV results
• Less than 2%
• Interesting seasonal
and geographic
variations
• High and low years not
always normal
distributed
Sites (N to S) IAV (14 years)
MRP 2 1.8%
MRP 1+ 1.3%
Pam. Camarones 1.1%
MRP 3 1.1%
Crucero 0.8%
Salar 0.9%
San P. de Atacama 1.5%
Puerto Angamos 1.8%
Cerro Armazones 0.4%
Salvador 0.8%
MRP 4 0.9%
Inca de Oro 0.8%
MRP 5 0.9%
MRP 6 1.4%+only 13 years of data available
Distribution of individual years around the long-term mean
0.8
0.85
0.9
0.95
1
1.05
1.1
1 2 3 4 5 6 7 8 9 10 11 12 13 14
No
rmal
ized
yea
rly
aver
age
GH
I
[1] MRP2 (IAV:1.8%)[2] MRP1 (IAV:1.3%)[3] P. Camar. (IAV:1.1%)[4] MRP3 (IAV:1.1%)[5] Crucero (IAV:0.8%)[6] Salar (IAV:0.9%)[7] SPd Ataca (IAV:1.5%)[8] P Angamos (IAV:1.8%)[9] C. Armaz. (IAV:0.4%)[10] Salvador (IAV:0.8%)[11] MRP4 (IAV:0.9%)[12] Inca de Oro (IAV:0.8%)[13] MRP5 (IAV:0.9%)[14] MRP6 (IAV:1.4%)
Conclusions
• IAV in the Atacama Desert significantly lower
than the 4 – 6% generic figure
• Less than 2% IAV estimate is subject to
uncertainties, longer ground measurements
would be desirable to verify findings in the
future
Thank you for your attention
Contact for further information [email protected]
Additional slides
-5%
-4%
-3%
-2%
-1%
0%
1%
2%
3%M
RP
2 (
1.8
%)
MR
P 1
(1
.3%
)
P.C
am.(
1.1
%)
MR
P 3
(1
.1%
)
Cru
cero
(0.8
%)
Sala
r (0
.9%
)
SPd
. Ata
. (1
.5%
)
P. A
nga
. (1
.8%
)
C. A
rma.
(0
.4%
)
Salv
ado
r (0
.8%
)
MR
P 4
(0
.9%
)
Inca
de
Oro
(0
.8%
)
MR
P 5
(0
.9%
)
MR
P6
(1
.4%
)
Dev
iati
on
fr.
lon
g-te
rm a
vera
ge (
%)
highest (%) lowest (%)
Relative deviation of years with lowest and highest GHI
IAV value stated in brackets behind the site name
Top Related