PACA assessing solar_radiation_data

High Spatial Resolution Solar Atlas In
Provence-Alpes-Côte d„Azur
(ISES Solar World Congress 2011)

Assessing Solar Radiation Data For
Large Scale Solar Energy Projects

Etienne Wey – TRANSVALOR S.A.

Introduction
2

 Cartography of Solar Resource at different scales
Worldwide yearly sum of global horizontal irradiation map
(kWh/m2) from NWP re-analysises (res. ~ 50 km)
3500 kWh/m²
80

60
3000 kWh/m²

40

20
2500 kWh/m²
High resolution yearly sum of GHI map (kWh/m2)
0 2000 kWh/m²
for the region Provence-Alpes-Côte d’Azur (res.
-20 200 m)
1500 kWh/m²
-40
Yearly sum of GHI map (kWh/m2)
for Europe from HelioClim databases
-60 1000 kWh/m²

(res. ~5 km, method Heliosat-2 applied to Meteosat images)
-80
500 kWh/m²
55
-150 -100 -50 0 50 100 150
2400

2200
50
2000

1800

45
1600

1400

40
1200

1000

35
-20 -10 0 10 20 30 40

Solar Atlas for the Mediterranean Stakeholder Workshop Nov. 1st, 2011

Introduction
3

 Project main characteristics:
 Based on HelioClim-3 satellite-based surface solar irradiation database
(based on Meteosat Second Generation: res. ~ 4 km, near real-time from 2004)
 200 m resolution solar maps
 Monthly and yearly sum of Irradiations
 Long-term mean and (standard deviation)
 Time series of monthly and yearly irradiations
 PV and Solar Thermodynamic Applications
 Global irradiations on typical tilted plans (e.g. for PV, thermal systems)
 Direct Irradiation on typical tilted plans and in normal incidence (e.g. for CSP and CPV)


Introduction
4

 The potential end-users and usages of a solar atlas:
 Governmental and private actors
 Geographical analysis of local solar potential
 Sitting and sizing solar power plants
Advanced feasibility pre-studies based on geographical analysis
(before, for example, the local installation of a pyranometer station)

 Individual
 High resolution map suitable for sizing small individual solar systems
(small PV system, solar water heating systems, etc)
 Accurate and well-presented solar maps are concrete and instructive for
everybody (e.g. education) to promote solar energy


Increase the spatial resolution of
HelioClim-3
5

 Intra-pixel (HC-3) effects of the relief
 Use of the relief database SRTM (Shuttle Radar Topography Mission)
 Spatial resolution of 100 m
 Localization Accuracy better than 10 m
 Effect of the optical depth variations of the atmosphere
(Abdel Wahab et al., 2008)
 Shadow effects respectively for the diffuse and direct components of the
global irradiation (Ruiz-Arias et al., 2010)

 Local calibration of irradiation estimation with on-ground
pyranometric measurements
 Calibration of the global horizontal irradiation (GHI)
 Calibration of the parametric experimental model of global/diffuse
decomposition
 Modeling of uncertainty from the calibration residue analysis


Shadows effects on solar
irradiation
6

 Example : Annot


The meteorological stations for
the HelioClim calibration
7

 29 Météo France ground stations
 Mostly hourly and daily GHI data

 The three meteorological stations
for the project (RSP)
 One year renting (CSP Service)
(spin-off of the German Aerospace, DLR)
 global and diffuse irradiations on
horizontal plan (sampling : 10 min)
(+ temperature and humidity)
 RSP : accurate system and robust
with respect to
 dust
 misalignments


« Raw* » HelioClim-3:
monthly GHI estimation errors
8 NDATA MREF MBE MAE RMSE CC
Monthly sums of GHI
1269 months 132.2 kWh/m2 5.3 % 7.5 % 10.1 % 0.992
(Reference: MF stations)

* “Raw” HC3:
• No correction of
orography effects
• No local calibration

RMSE

MBE <0

MBE >0


Corrected* HelioClim-3:
monthly GHI estimation errors
9 NDATA MREF MBE MAE RMSE CC
Monthly sums of GHI
1269 months 132 kWh/m2 -0.2 % 3.1 % 4.3 % 0.996
(Ref: 20 MF stations)
Monthly sums of GHI
LOOCV 1267 months 132 kWh/m2 -0.7 % 4.4 % 6.2 % 0.991
(Ref: 20 MF stations)
Monthly sums of GHI
38 months 126 kWh/m2 0.4 % 2.5 % 3.1 % 0.998
(3 RSP stations)

RMSE dependency with orography
Mountainous area:
5.8 % => 7.4 kWh/m2 (et MBE = -0.6%)
Non-mountainous area:
3.2 % => 4.3 kWh/m2 (et MBE < -0.1%)
* Corrected HC3:
• Orography effects
• Local calibration

RMSE

MBE <0

MBE >0


Example of high resolution
irradiation map
10 Map of yearly sums of GHI (mean between 2004 –2010)


Example of high resolution
irradiation map
11 Map of yearly sums of DNI (mean between 2004 –2010)


Applications from the solar atlas
12

 Ranking maps to determine “best” potential solar sites

SOLAR
ATLAS PACA

« Geophysial »
conditions analysis
Exclusion maps Best
creation
potential
Electrical grid analysis
(source points « Ranking maps »
sites
localisation) creation identification


Exclusion maps
13

• Lakes
• Floodplain

• Agriculture, forests, cities,…
• Natural preservation zones

• Slope over 15% (or 20% if south oriented)


Exclusion maps
14

 35% of the initial surface
remains after application of
the exclusion zones
 Natural preservation zones
 Forests
 Floodplains
 High slope values


Ranking maps – Criteria 1
15

 Distance from grid source points
Value (km) Mark
<à 1,0 20
<à 2.6 19
<à 4.2 18
<à 5.8 17
<à 7.4 16
<à 9,0 15
<à 10.6 14
<à 12.2 13
<à 13.8 12
<à 15.4 11
<à 17,0 10
<à 18.6 5
<à 20.2 4
<à 21.8 3
<à 23.4 2
<à 25,0 1
<à 26.6 0

16

 Mean yearly irradiation for south oriented panel with 35° slope
Value Rank
>à 2200 20
>à 2125 19
>à 2050 18
>à 1975 17
>à 1900 16
>à 1825 15
>à 1750 14
>à 1675 13
>à 1600 12
>à 1525 11
>à 1450 10
>à 1375 9
>à 1300 8
>à 1225 7
>à 1150 6
>à 1075 5
>à 1000 4
>à 925 3
>à 850 2
>à 775 1
>à 700 0


17

 Nature parks

Value Rank
Natural park 10
Other 20


Ranking maps – Final
18

 Final global ranking map including:
 Distance to source points
 Irradiation
 Nature parks


Potential best sites ranking map
19

 Final map to assess the “best” sites potential:


Finalising the solar radiation data
assessment for a specific site
20

 Installation of an irradiation measurement ground station on site for 6
to 12 month (need to “catch” the maximum variation during the year)
 Calibration of the satellite irradiation data from the measure
 Creation of Typical Meterological Year time series (P50, P90) to
estimate the solar energy project yield
 Verify the site irradiation variability if larger than satellite pixel size


Conclusion - 1
21

 High resolution Solar Atlas
 Resolution: 200 m
 HelioClim-3 database (res. 4 km, period: 2004 - 2010)
 Shadow effects estimated from the DEM SRTM
 Local calibration with ground pyranometric stations
 Monthly sum of global and direct irradiation on different tilted plan
 Uncertainties estimated from statistical analysis of the calibration residue
w.r.t. pyranometric ground stations
 Monthly sum of global irradiation: bias < 1 %, RMSE ~ 5 % (~7 kWh/m2)
 Monthly sum of direct normal irradiation: bias < 1 %, RMSE ~8 % (~12.5
kWh/m2)


Conclusion - 2
22

 Creation of a potential “best” sites map
 Exclusion zones based mainly on land cover and Digital Elevation Model
 Ranking maps based on :
 Distance from electricity grid source points
 Yearly mean irradiation value (specific to the technology of interest – PV, CPS, CSP)
 Societal acceptability of the zone (Natural park,…)

 Fine local assessment for the site
 Installation of an irradiation measurement ground station
 Calibration of the satellite data with the ground station
 Creation of TMY time series (P50, P90) to calculate the solar energy project
yield
 Estimation of the local variation of irradiation is site is large


PACA assessing solar_radiation_data

Recomendados

Recomendados

Más contenido relacionado

La actualidad más candente

La actualidad más candente (20)

Similar a PACA assessing solar_radiation_data

Similar a PACA assessing solar_radiation_data (20)

Más de RCREEE

Más de RCREEE (20)

PACA assessing solar_radiation_data