|
|
|
Clean Energy
Institute
College of Engineering, Technology, and
Architecture
University of Hartford |
|
|
Comparison Between Predicted and Measured Power of Solar Insolation for
a series of Photovoltaic Arrays
|
|
Problem
Statement: There are a number of models
for predicting solar insolation at specific
locations. Variations of these models are used size
solar photovoltaic arrays based on predictions of
the electrical energy produced over the course of an
average year. The problem is to compare predictions
based on two of these models with measured values of
solar insolation and predictions of electrical
energy when compared to values recorded for a
specific array.
Description:
A set of four arrays, consisting of 12 sets of
photovoltaic cells per array, for a total of 48
cells, covering 46.45 m2, are located in
Tolland, Ct. Software provided with the inverters
record electrical energy produced over time. The
model proposed by Goswami,
et al, and the Clear Sky Model
,recommended by ASHRAE, are used to predict solar
insolation and energy generated, for specific days
during the month of August, 2005. Solar insolation
measurements, using a light meter calibrated in
Watts/m2. Values of percentage of sun,
based on Typical Meteorological Years
data, complied by NREL, are used to calculate
energy, for the same period of time, and compared
with predicted values. The comparison shows the
Goswami model over predicts both insolation and
energy. The ASHRAE model, while indicating better
agreement for solar insolation, tends to overpredict
the energy produced. This difference is thought to
be due to shadowing of portions of the arrays over
the course of the day and atmospheric conditions;
e.g., high humidity, haze, etc, as well as possible
variations of cell efficiency with temperature over
the course of a day and from day to day. Using these
results, predictions for a typical year indicate
approximately 23% more energy than recorded.
University
of Hartford Participants: |
|
Barry O’Grady
Faculty Advisor: Barry Lubin |
|
|
|
|
Client: Barry
Lubin, PhD | | |
|
| |
