Range limit check for sunshine duration

1
Improvements of the quality control system for
solar radiation and sunshine duration data at
the Centre of Meteorological Experimentations
(ReSMA) of the Italian Air Force National Met
Service (INMS)
S. Vergari, F. Foti, E. Vuerich, P. Cucchiarelli
and T. Iorio
ReSMA - Centre of Meteorological
Experimentations, via Braccianese Claudia Km
20.100, 00062 Bracciano, Italy. Tel. 39 06
9988 7701 Fax [phone] E-mail
stefania.vergari_at_aeronautica.difesa.it vergari,
foti, vuerich_at_meteoam.it
Introduction
Solar radiation and sunshine duration are two of
the most important variables in the energy budget
of the Earth. They play a fundamental role in the
performance estimation of renewable energy
systems and in many other applications
(ecological building designs, agriculture, human
healthy). An effective system of quality control
therefore need to ensure that only acceptable
data could be collected and data meet the quality
requirements for the intended application. In
order to improve quality control checks for
global solar radiation and sunshine duration data
of the INMS network, we implemented two different
quality controls. A range limit check, applied to
both variables separately, concerns the respect
of variables physical limits and has been
improved varying physical limits in agreement to
the latitude and the season. An internal
consistency check, applied for the first time in
2009, provides a physically real area when both
measures are available simultaneously. From the
climatic history of each site, by applying the
linear form of the Angstrom-Prescott model, the
atmospheric clearness index has been monthly
calculated. Then an upper and a lower bound for a
value of solar radiation are defined as linear
functions of this index and the sunshine
duration.
The Centre of Meteorological Experimentation of
Vigna di Valle (420449 N 121241 E 266 m
a.s.l.), collects solar data from the INMS solar
network. Solar radiation and sunshine duration
data is checked by a quality control system and
delivered to the World Radiation Data Center
(WRDC) of St. Petersburg (Russia).
R amount of measured daily solar
radiation Ro extraterrestrial daily solar
radiation S number of sunshine hours daily
recorded So maximum day length a,b regression
constants monthly determined
CM 11 KZ
Method
Results
Range limit check for sunshine duration
Year 2009
If S(X,D) sunshine duration (hours) of the day
D at the meteorological station X
Amount of daily data checked
18000
0 S(D,X) So(X,D)
where So(X,D) maximum day length of day D at
the latitude of X
Detected errors 1.8
Range limit check for global solar radiation
Corrections done 0.5
If R(X,D) Global solar radiation (MJ/m2) of
the day D at the meteorological station X
Erased data 1.2
0 R(D,X) lt Ro(X,D)
False detections 0.1
where Ro(X,D) extraterrestrial radiation for
day D at the latitude of X
Data recovery not achievable because of mainly
concerns solar radiation data corrupted by
instrumentation breakdown.
Internal consistency check
Given S(X,D)
LB(K, S(X,D)) R(D,X) lt UB(K, S(X,D))
Where functions LB(K, S(X,D)) and UB(K,
S(X,D)) depend on the clearness index K (monthly
determined site by site) which aims to consider
the availability of solar radiation incoming in
the atmosphere. In UB, K lessens the
extraterrestrial upper limit Ro(X,D), according
to the hours of sunshine duration recorded. In
LB, K heightens the lower limit of direct solar
irradiance equal to 120 W/m2 multiplied by the
hours of sunshine duration recorded.
In the graphical example
2 errors detected
UB
that, at manual inspection on row data,
resulted
LB
2 real errors on sunshine duration values
The control quality system is implemented by a
MATLAB code it takes as input daily or hourly
values for both variables and gives as output
monthly reports of detected errors (graphics
production is optional). A table of clearness
indexes K, monthly computed for each solar
station, is also required in order to compute UB
and LB. Regression constants of the
Angstrom-Prescott model, needed to compute K, are
derived from historical series of solar variables.
Slopes of UB and LB vary both site by site and
monthly
Conclusions and future work
The purpose of this work is to realize a
selective quality control for global solar
radiation and sunshine duration data, involving a
range limit check together with an internal
consistency check working on both hourly and
daily data. The aim is to perform a finer control
on data reducing as much as possible the number
of false errors detected. In the near future,
the increasing number of automatic weather
station, provided with high performance
instrumentation, will imply a quickly grow of the
amount of data to manage a finer automatic
quality control system is therefore needful to
ensure, to the scientific communities, the
respect of international quality standards.