Solar resource measurements and sattelite data

Solar Resource Measurements
and Satellite Data
4th Sfera Summer School 2013
Hornberg Castle, Germany
Dr. Norbert Geuder
CSP Services – Almería – Cologne

Outline
1) Introduction
2) Fundamentals on Solar Irradiation
3) Variability of Irradiation
4) Irradiation Sensors
5) Sensor calibration and measurement accuracy enhancement
6) Satellite Based Assessment
7) Outlook / Summary

4th Sfera Summer School, Hornberg Castle, 2013

-2

General Procedure at Solar Resource Assessment
Irradiation map: spatial distribution

Geographical data: land use, etc.

+
GIS analysis

Selection of promising sites

Meteorological Station:
Accurate irradiation data
Long-term time series
(>10 years)
Plant design, inter-annual variability, uncertainty analysis, financing, …

-3

Solar Resource Assessment for CSP Plants
• Direct Beam Irradiation data required for CSP Applications

not global irradiation – this is quite a difference!
• Usually not available in suitable sunny regions in the world
• Accessible via measurements or derived from satellite data
Restrictions:
– Measurements: expensive, long duration, not for the past
– Satellite data:

high uncertainty (≈ 10 % or more)

• High accuracy required for DNI with long-term performance


-4

Impact of Solar Resource Uncertainty
on CSP Plant rentability
DNI
uncertainty

expected long-term value (100%)

(e.g. ±10 %)

Annual DNI

Electricity
production

Earnings
Losses

Annual
expenses
(redemption,
O&M, …)

Irradiation uncertainty decides over project realization !
With thoroughly performed measurements an accuracy of approximately 2 % is achievable.

-5

Fundamentals on Solar Irradiation


-6

Solar Constant

1367 W/m²
Calculation:

L
W
 1367 2
4r 2
m

Luminosity of the sun: L = 3.86 x 1026 W
Astronomical unit:

r = 149.60×109 m

Annual Variation of Solar Constant
1420

Irradiance Outside Atmosphere (W/m²)

Mean solar irradiance (flux density
in W/m²) at normal incidence
outside the atmosphere
at the mean sun-earth distance r.

1410
1400
1390
1380
1370
1360
1350
1340
1330
1320

0

90

180

270

360

Day of Year


-7

Path of Solar Radiation through the atmosphere
Radiation at the top of atmosphere

Ozone.……….…....

Absorption (ca. 1%)

Air molecules..……

Rayleigh scattering and absorption (ca. 15%)

Aerosol…….………..…...……

Scatter and Absorption ( ca. 15%, max. 100%)

Clouds………….………..

Reflection, Scatter, Absorption (max. 100%)

Water Vapor…….……...………

Direct normal irradiance at ground

Absorption (ca. 15%)

Source: DLR


-8

Radiative Transfer in the Atmosphere
1400

Extraterrestrial
O2 and CO2
1000

Ozone
800

Rayleigh
600

Water Vapor
400

Aerosol
200

Clouds

Hour of Day

00:00

22:00

20:00

18:00

16:00

14:00

12:00

10:00

08:00

06:00

04:00

02:00

0
00:00

Direct Normal Irradiation
(W/m²)

1200

Source: DLR


-9

Air Mass
AM = 0 (outside of atmosphere)

Source: DLR


- 10

Solar Spectrum and Atmospheric Influence
1 Planck curve T=5780 K at mean sunearth distance
2 extraterrestrial solar spectrum
3 absorption by 03
4 scattering by 02 und N2
5 scattering by aerosols
6 absorption by H2O vapor
7 absorption by aerosols

UV radiation:

0.01 - 0.39 µm, ~ 7 %

Visible Spectrum: 0.39 – 0.75 µm, ~ 46 %
Near infrared:

0.75 – 2.5 µm, ~ 47 %

www.volker-quaschning.de/articles/fundamentals1/index.php


- 11

Characteristics of solar irradiation data
•

Component:
– DNI (Direct-Normal Irradiation)
– DHI (Diffus-Horizontal Irradiation)
– GHI (Global-Horizontal Irradiation)

•

Source:
– ground measurements:
• precise thermal sensors (thermopiles)
• Rotating Shadowband Irradiometers (RSI)
– satellite data

•

Properties of irradiation:
– spatial variability
– inter-annual variability
– long-term drifts

- 12

Direct, Diffuse and Global Irradiance
When measuring solar irradiance, the
following components are of particular
interest:

Zenith angle θ,
solar elevation 

 Direct normal irradiance (DNI)
(also: beam irradiance)

 Diffuse horizontal irradiance (DHI)
(also: diffuse sky radiation)

 Global horizontal irradiance (GHI)
(also: total solar irradiance)



GHI = DHI + DNI * sin ()

- 13

Direct Normal Irradiation (DNI)

DNI = BHI / sin 
with: BHI = Beam Horizontal Irradiation
direct

Example:
BHI = 600W/m²
 = 50°
 DNI = 848W/m²
DNI > BHI



Direct-Normal- Irradiation (DNI)


- 14

Global Horizontal Irradiation (GHI)

GHI = BHI + DIF
Example:

direct
diffuse

BHI = 600W/m²
DIF = 150W/m²
 GHI = 750W/m²

diffuse

Global-Horizontal-Irradiation (GHI)


- 15

Long-term variability of solar irradiance

GHI from Potsdam, Germany


- 17

Long-term variability of solar irradiance

Source: DLR

7 to 10 years of measurement to get long-term mean within 5%

- 18

Inter-annual variability
1999

deviation
to mean

2000

kWh/m²a

Average of the direct normal
irradiance from 1999 to 2003

Strong inter-annual and
regional variations

2001

2002
Source: DLR

2003

- 19

Irradiation sensors


- 20

Absolute Cavity Radiometer
Principle of Measurements:
-

-

Possibility to measure absolute irradiance
values. All other irradiance measurement
devices need to be calibrated using an absolute
cavity radiometer
Its principle of operation is based on the
substitution of radiative power by electrical
(heating) power

-

Measurement in intervals with minimal length
of 45 s. Constant irradiation required for
measurement campain

-

Tracking device required

-

No continuous measurement (!)

ftp.pmodwrc.ch/pub/pmo6-cc/user_guide_11.pdf

Valid for calibration purposes


- 21

Suitable equipment for irradiance measurements for
Concentrating Solar Power (CSP)
Thermal sensors

Semiconductor sensor
Rotating Shadowband Irradiometer,
RSI
(photodiode)

Pyranometer,
pyrheliometer
(thermopiles)

- 22

Thermopile Sensors
Shading assembly with shading ball

CMP21 Pyranometer (GHI, DHI shaded)
with ventilation unit CVF3

CHP1 Pyrheliometer (DNI)

Sun sensor
Solys 2 sun tracker


- 23

Thermopile Sensors – Pyrheliometer
Principle of Measurement:
-

Pyrheliometer = radiometer suitable to
measure direct normal irradiance

-

Highly transparent window 97 – 98 %
transmission of solar radiation

-

Housing geomerty with 200 mm absorber tube
restricting acceptance angle to 5°

-

Sensing element with black coating and
built-in termopile device

-

www.kippzonen.com/?product/18172/CHP+1.aspx

Pt-100 temperature sensor for temperature
corrections


- 24

Pyrheliometer Specifications

Kipp&Zonen CHP 1 Specifications:
Spectral range: 200 to 4000 nm
Sensitivity: 7 to 14 µV/W/m² (mV/kW/m²)
Response time: < 5 s
Expected daily uncertainty: ± 1 %
Full opening view angle: 5° ± 0.2°
Required tracking accuracy: ± 0.5° from
ideal



- 25

Thermopile Sensors – Pyranometer
-

Pyrheliometer = radiometer suitable to
measure short-wave (0.2 - 4 µm)
global or diffuse radiation

-

Highly transparent glass dome 97 – 98 %
transmission of solar radiation

-

Full view on 2π hemisphere
(horizontal levelling required)

-

Sensing element with black coating and
built-in termopile

-


Pt-100 temperature sensor for temperature
corrections


- 26

Thermopile Sensors – Pyranometer
Kipp&Zonen CMP21 Specifications:
Spectral range: 285 to 2800 nm
Sensitivity: 7 to 14 µV/W/m² (mV/kW/m²)
Response time: 5 s

www.kippzonen.com/?product/1491/CMP+21.aspx


- 27

Rotating Shadowband Pyranometer (RSP)
Licor silicon photodiode

Rotating shadowband

Housing of mechanics


- 28

RSI – Principle of Measurement

Simplified sensor signal during shadow band rotation:
once per minute, rotation lasts about 1.5 seconds

Source: Solar Millennium AG

- 29

Licor Li-200 Pyranometer Sensor

Specifications:
Sensitivity:
Typically 90 µA per 1000 W/m²
Response time: 10 µs.
Spectral range: 0.4 – 1.1 µm
Calibration:
Calibrated against an Eppley Precision
Spectral Pyranometer under natural
daylight conditions.
Typical error under these conditions is
±3% up to ±5%.
www.licor.com/env/Products/Sensors/200/li200_description.jsp


- 30

Precise thermal sensors:

Pyrheliometer and Pyranometer on sun tracker
Advantages:
+ high accuracy (1 to 2%)
-GHI

-DNI

-DHI

+ separate sensors for
GHI, DNI and DHI
(cross-check through redundancy)

Disadvantages:
- high acquisition costs
- high maintenance costs
- high susceptibility for soiling
- high power demand
(grid connection required)

- 31

Sensor with photo diode:
Rotating Shadowband Irradiometer, RSI
Advantages:
+ fair acquisition costs
+ low maintenance
+ low susceptibility for soiling
+ low power demand (PV-Panel)

Disadvantage:
- reduced accuracy due to systematic
deviations of the photodiode sensor
response:
primordial DNI: ≈ 6 to 10 %
(or even higher)

- 32

Measurement uncertainty

Precise instruments (HP) versus RSI
Error source:

Pyrheliometer:

RSI:



Calibration

< ±1.1%

±3% (... ±5%)



Temperature

< ±0.5%

0% ... ±5%



Linearity

< ±0.2%

±1%



Stability

< ±0.5%/a

< ±2%/a



Spectral dependence

<±0.1%



Sensor soiling

-0.7% per day

0% ... ±8%
-0.07% per day

systematic errors

can be corrected!!

- 33

Choice of Measurement Equipment
Which equipment is suitable for measurements in Solar Resource Assessment?

?
High Precision sensors (thermopiles)

Rotating Shadowband Irradiometer:
RSI

- 34

Objectives for Irradiance Measurements
Solar Resource Assessment

Power Plant Monitoring

•

at remote site

•

no qualified staff

•

always qualified staff on site

•

no electric grid

•

electric power available

•

often dusty and arid areas


- 35

Pyrheliometer soiling in southern Spain

Plataforma Solar de Almería

- 36

Comparison of sensor soiling

University of Almería

- 37

Soiling characteristics of pyrheliometers and RSI‘s
Solar
Irradiation

direct
sunlight

glass plate
tube with
200 mm
length

diffusor disk
over
photodiode

absorber

Pyrheliometer

RSI
sensor head

- 38

Choice of the adequate equipment
For Solar Resource Assessment
• at remote sites and
• daily maintenance not feasible
an RSI is the premium choice for DNI measurements.
However:
• Proper calibration
• Corrections of systematic signal response
• regular maintenance inspections (2 to 4 weeks)
are indispensable for reliable measurements.


- 39

Sensor calibration
and measurement accuracy enhancement


- 40

Sensor Calibration – Fundamentals I
- The World Standard Group
(WSG) is an assembly of highly
precise absolute cavity
radiometers.

www.pmodwrc.ch/pmod.php?topic=wrc

- The measured mean value
(World Radiometric Reference)
is the measurement standard
representing the SI unit of
irradiance with an estimated
accuracy of 0.3 %.
- All other short wave irradiation
measurement systems are
calibrated against this single
value.

Precision measurement at the
WorldRadiation Center (WRC)


- 41

RSI sensor calibration by DLR on PSA

2-monthly calibration of each RSI against
high-precision instruments
at Plataforma Solar de Almería (PSA)
(recommended every 2 years)

- 42

RSI sensor calibration duration

Variations of the Calibration Constant with calibration duration

DNI


- 43

Variability of the Correction Factors (CF)

Variability of
correction factors
(CF):
radiation components
need to be corrected
with separate CFs.


- 44

Recalibration of RSIs

Drift of Calibration Factor within 2 to 4 years


- 45

Correction of raw RSI measurement values
Origin of systematic errors of RSI response
• Temperature dependence of semiconductor sensor
• Spectrally varying irradiation
– different for irradiation components (direct beam / diffuse)
– depending on Air Mass
• Angle of incidence

• Pre-calibration of the sensor head (from the manufacturer)


- 46

Spectral correction of diffuse irradiation
raw values

variation with
air mass + altitude
corrected

Π spec 

DNI  GHI
DHI 2


- 47

Dependence of response on solar elevation

BHIref / BHIRSI

so called „cat-ear effect“

solar elevation angle in degree

Correction applied only on direct beam portion of the global response

- 48

BHIref / BHIRSI

Dependence of response on solar elevation

corrected

solar elevation angle in degree


- 49

Reachable accuracy for DNI with RSIs
Accuracy of RSI
measurements
as derived from a comparison
of the data from 23 RSIs

RMSD = 13 W/m²

with precise thermopile
measurements
within the course
of a whole year

GHI

DHI

DNI

raw

cor

raw

cor

raw

cor

reference
(CHP 1)

average MB

-10.3 ± 4.0

0.3 ± 1.3

-17.3 ± 1.6

-0.4 ± 0.7

24.6 ± 10.5

1.0 ± 0.5

1.0 ± 3.9

W/m²

RMSD

14.2

7.6

18.9

4.5

33.3

13.0

5.3

W/m²

Annual sum

up to -2.5

< ±1

up to -15

< ±3.5

up to +7

< ±1

up to 1.3

%

RSP

unit

10 min time resolution

- 50

Transferability of the results?
The reachable accuracy of the measured beam irradiance data
for the client at his prospected sites depends on 2 crucial points:
• Stability of the sensor sensitivity
• Transferability of the results to other sites
• Regular inspections and data controlling


- 51

Transferability to other sites and climates
Parallel measurement campaigns
in UAE:
• Comparision of 6 RSIs to high-precision
thermal sensors
• Measurement periods >3 weeks
• in summer and winter


- 52

Relative deviation of DNI sum within measurement
campaign
only summer:
DNI < 730 W/m²
summer + winter:
DNI until 1000 W/m²


- 53

Satellite Based Assessment


- 54

Satellite Derived Data

www.solemi.de/method.html

Analyze satellite data in two steps:
1. Atmosphere: Gather satellite information of
atmospheric composition (ozone, water vapor
and aerosols) and apply the ‘clear sky model’ to
calculate the fractions of direct and diffuse
irradiance
1400

O2 and CO2
1000

Ozone
800

Rayleigh
600

Water Vapor
400

Aerosol
200

Clouds
00:00

22:00

20:00

18:00

16:00

14:00

12:00

10:00

08:00

06:00

04:00

02:00

0
00:00

Direct Normal Irradiation
(W/m²)

2. Clouds: Calculate the cloud index as the
difference between actual reflectivity of the
earth as it is seen by the satellite and a
reference image which only includes
reflectance of the ground

Extraterrestrial
1200

Hour of Day


- 55

How to derive irradiance data from satellites

 The Meteosat satellite is
located in a
geostationary orbit
 The satellite scans the
earth line by line every
half hour

Scan

Source: DLR


- 56

How to derive irradiance data from satellites

Scan in visible spectrum

Derivation of a
cloud index
from the two
channels

Scan in infra-red spectrum

Source: DLR


- 57

Different Cloud Transmission for GHI and DNI
 Global Irradiation

 Direct Irradiation
Sun-satellite angle 60-80

1.2

1.2

-26 °C
-16 °C
-6 °C
4 °C
14 °C
-30°C - -20°C
-20°C - -10°C
-10 °C - 0°C
0°C - 10 °C
>10°C

1

cloud transmission

1
0.8

0.8

0.6

0.6

0.4

0.4

0.2

0.2

0

Different exponential
functions for varying
viewing angles and
brightness
temperatures

0

-0.2
-0.2

0

0.2

0.4

0.6

cloud index

0.8

1

1.2

-0.2
-0.2

0

0.2

0.4

0.6

0.8

1

1.2

Source: DLR


- 58

Clear sky Model input data
 Aerosol optical thickness
 GACP Resolution 4°x5°, monthly data
 MATCH Resolution 1.9°x1.9°, daily data
 Water Vapor:
NCAR/NCEP Reanalysis
Resolution 1.125°x1.125°, daily values
 Ozone:
TOMS sensor
Resolution 1.25°x1.25°, monthly values
Source: DLR

- 59

Uncertainty in Aerosols

GADS

Toms

GOCART

NASA GISS v1 / GACP

All graphs are for
July
Scales are the same!
(0 – 1.5)
Large differences in
Aerosol values and
distribution

NASA GISS v2 1990

AeroCom

Linke Turbidity
Source: DLR

- 60

Comparing ground and satellite data:
“sensor size”

solar thermal
power plant
(200MW  2x2
km²

satellite pixel
( 3x4 km²)

ground
measurement
instrument
(2x2 cm²)

Source: DLR


- 61

accuracy
general difficulties:
• point versus area
• time integrated versus
area integrated
1200

ground
1000

satellite

W/m²

800
600
400
200
0
0

6

12

hour of day

18

24

Source: DLR


- 62

time scales

12:45

13:00

13:15

Hourly average

13:30

13:45

Meteosat image

14:00

14:15

Measurement

 Ground measurements are typically
pin point measurements which are
temporally integrated

Hi-res satellite pixel in Europe

 Satellite measurements are
instantaneous spatial averages
 Hourly values are calculated from
temporal and spatial averaging
(cloud movement)

Source: DLR


- 63

Results of the satellite-based solar assessment
Digital maps: e.g. annual sum of direct normal irradiation
in 2002 in the Mediterranean Region

The original digital maps can
be navigated and zoomed with
Geographical Informations
Systems like ArcView or Idrisi.

data produced by

(DLR, 2004) for MED-CSP

Temporal resolution of input data: 1 hour
Spatial resolution of digital map: 1 km x 1 km per Pixel
Long term analysis: up to 20 years of data

Source: DLR


- 64

Results of the satellite-based solar assessment
Time series: for single sites, e.g. hourly, monthly or annual
Hourly DNI [Wh/m²] for one site in Spain

Annual sums of DNI [kWh/m²] for one site in Spain

Monthly sums of DNI [kWh/m²] for one site in Spain

Hourly monthly mean of DNI in Wh/m², Solar Village 2000

hour

Source: DLR


- 65

Satellite data and nearest neighbour
stations
Satellite derived
data fit better
to a selected
site than
ground
measurements
from a site
farther than
25 km away.

Source: DLR

- 66

Ground measurements vs. satellite derived data
Ground measurements

Satellite data

Advantages

Advantages

+ high accuracy

(depending on sensors)

+ high time resolution

+ spatial resolution
+ long-term data

(more than 20 years)

+ effectively no failures
+ no soiling
+ no ground site necessary
+ low costs

Disadvantages
- high costs for installation and O&M

Disadvantages

- soiling of the sensors

- lower time resolution

- possible sensor failures

- low accuracy at high time resolution

- no possibility to gain data of the past

- 67

Combining Ground and Satellite Assessments
• Satellite data
– Long-term average
– Year to year variability
– Regional assessment
• Ground data
– High Precision (if measurements taken thoroughly)
– High temporal resolution possible
(up to 1 min to model transient effects)
– Good distribution function
– Site specific


- 68

Procedure for Matching Ground and Satellite Data
Ground
measurements

Recalculation

Separation of clear sky
and cloud conditions
clear sky correction

Comparison

with alternative cloud
transmission tables

Recalculation

Selection

with alternative
atmospheric input

Selection
of best atmospheric input

of best cloud
transmission table

cloud correction

Satellite
assessment

Transfer MSG to MFG
Recalculation of
long term time series

Best fit
satellite data


- 69

What you should care for in good
Solar Resource Assessment


- 70

Procedure to Follow for Proper Solar Resource
Assessment
•

Find a good location: close to site, safe, suitable for collocation of Weather
Station

•

Clarify the ground property conditions

•

Check/define the budget for:
instrumentation, maintenance and measurement related services

•

Select the appropriate measurement equipment and provider
(based on budget considerations, local conditions on site and maintenance
possibilities)

•

Find local maintenance personnel

•

Prepare the measurement site according to the supplier’s specifications
(foundations, fencing, etc.)

•

Installation and commissioning of the measurement equipment

•

Steady monitoring of the measurement data,
duration minimum 1 year

- 71

Procedure to Follow for Proper Solar Resource
Assessment
•

Documenting the selection of instruments

•

Choosing a renowned company or institution to conduct or assist the
measurement campaign

•

Documenting sensor calibration with proper calibration certificates

•

Meticolously documenting the instrument installation and alignment

•

Performing and documenting regular sensor cleaning, maintenance and
verification of alignment

•

Cautiously and continuously checking data for errors and outliers

•

Flagging suspect data, and applying corrections if possible, during and after
the measurement campaign

•

Stating and justifying the uncertainty estimate in a detailed report after the
measurement campaign.


- 72

Usual Expert Service for Solar Resource Assessment
Expert office

On site
Daily data retrieval via
modem (GSM/GPRS)

Data collection and processing:
•
•
 Installation & commissioning
 Operational supervision and
control
 Equipment monitoring
with inspection visits on site

Client

quality and functionality check

•

 Delivery of hardware

accuracy enhancement
(correction)
graphical visualization
Daily, monthly,
annual report with
good quality data
to client (via e-mail)


- 73

Quality Control of Measurement Data
 Are values physically possible ?
Measurement values must met
physical limits
 Are they reasonable?
e.g. comparison to a clear sky
model (Bird) or in kd-kt-space
 Are they consistent?
Comparison of redundant
information
 Visual inspection by an expert


- 74

Summary
• Knowledge of accurate irradiation data is indispensable for CSP
projects
(→ proper plant design, financial calculation, efficient plant operation)
•

site selection, pre‐feasibility with satellite data

•

colocation of a measurement station, taking care on thorough operation

•

match long‐term satellite data with good quality measurement data from ground

•

monitor the operating plant efficiency thoroughly with high‐precision data


- 75

Thank you
very much
for your
attention!


- 76

Solar resource measurements and sattelite data

Recommandé

Recommandé

Contenu connexe

Tendances

Tendances (20)

En vedette

En vedette (20)

Similaire à Solar resource measurements and sattelite data

Similaire à Solar resource measurements and sattelite data (20)

Plus de SolarReference

Plus de SolarReference (9)

Dernier

Dernier (20)

Solar resource measurements and sattelite data