Geology Reference
In-Depth Information
Chapter 5
Data Based Solar Radiation Modelling
Abstract This chapter describes the data based solar radiation modelling results
obtained from a case study. For the purpose or comparison of different data pre-
processing, data selection and data modelling approaches, the data (6-hourly
records and daily) from the River Brue catchment has been used. The Gamma Test,
Entropy theory, AIC (Akaike
s information criterion)/BIC (Bayesian information
criterion) have been explored with the aid of a nonlinear model LLR, ANFIS and
ANNs utilizing 6-h records in
'
first few sections of results and discussions. Later
modelling has been performed with Gamma Test and other nonlinear intelligent
models and other wavelet conjunction models on daily data from the Brue catch-
ment. Towards end of this chapter, we performed the best and useful data modelling
approach for the daily solar radiation modelling at the Brue catchment in terms of
very simple overall model utility comparison.
5.1 Introduction
As an important driving force that influences water circulation on earthen surface,
solar radiation plays an important role in hydrology. Solar radiation modelling has
great signi
nd
a lot of literature presenting solar radiation forecast and modelling based on several
data based statistical, fuzzy logic and neural network approaches around the globe.
Chen et al. [
13
] proposed ANNs and ANFIS based approaches to tell the difference
of the solar radiations between the different sky conditions. Takenaka et al. [
19
]
applied ANNs on solar radiation modelling based on radiative transfer along with
an improved learning algorithm to approximate radiative transfer code. Ahmed and
Adam [
7
] explored the possibility of ANNs to estimate monthly average daily
global solar radiation in Qena, upper Egypt; similar studies can
cance in this era of development of low-energy society. One could
find in other parts of
the world by means of meteorological parameters in Gusau, Nigeria [
1
], Mada-
gascar [
16
], United Kingdom [
6
,
17
]. Hybrid Models of ANN and SVM also
applied to Hydrology. Chen and Li [
14
] applied support vector machine for
