INTRODUCTION
Hybrid generation systems increase the reliability and sustainability of off-grid
power generation systems. In Malaysia, the setting up of the 100 kW solar, wind
and diesel hybrid system in Pulau Perhentian Kecil is an example of off-grid
power generation system. By integrating the solar and wind generator with the
diesel generator, the reliability of electricity generation in the island was
increased (Darus et al., 2009). Sizing optimization
for hybrid solar and wind power generation system ensures the system built is
with minimum cost and able to meet the load demand. The sizing analysis explained
in this study uses an iterative technique to calculate the size of components
such as PV panels, wind turbines, charge controllers, batteries and inverters
to be used in the hybrid generation system design. System reliability and system
cost is the most important parameters that need to be taken into account in
sizing optimization for hybrid PV and wind generation systems (Kaabeche
et al., 2011; Zhou et al., 2010).
There are several other optimization techniques for hybrid PV and wind generation
systems such as dynamic programming, graphical construction techniques, multi-objective
designs and artificial intelligence methods (Zhou et
al., 2010). A dynamic programming model sizing optimization technique
presented by Musgrove (1988) determines the optimal
operating strategies for a hybrid PV and wind generation system and an auxiliary
diesel generator. The techniques takes into account the average daily cost to
satisfy an electric load profile. Another hybrid PV and wind system optimization
method presented by Kaabeche et al. (2011) uses
an iterative optimization technique to follow the deficiency of power supply
probability model and the levelised unit electricity cost model for system power
and system cost reliability.
The objectives of this study are to develop a software that can optimize hybrid
solar and wind electricity generating systems and to build an optimum sized
hybrid solar and wind generator for sustainable electricity generation. The
project objectives will be achieved by performing several experiments to obtain
data for the software to be designed. A single axis tracker and a dual axis
tracker were built to compare the electricity generated from both the systems.
The hybrid generator sizing software will increase accuracy of sizing analysis
for hybrid wind and solar generators. The reason is the availability of solar
and wind resources at the location where the generator will be used will be
obtained for 7 days on hourly basis to get the weekly average and included in
the sizing analysis. Total load demand also will be observed on hourly basis
to obtain the maximum and minimum value and also the weekly average value and
included in the sizing analysis.
THEORETICAL DEVELOPMENT
The hybrid generating system design will consist of the following components
that can be seen in Fig. 1 (Kaabeche
et al., 2011). The sizing software developed will run several analysis
to make sure the components used in the hybrid generating system design is in
the optimum size taking into account the total load to be powered and also wind
and solar resource data at the region where the system will be used. The steps
to calculate size of all the components in the generator can be seen in the
flow chart in Fig. 2. The analysis performed by the software
can be divided into 8 parts as the following:
| • |
Resource data analysis |
| • |
Total daily load demand and corrected load demand analysis |
| • |
Battery size analysis |
| • |
Power generation ratio analysis |
| • |
Solar panel power rating analysis |
| • |
Wind generator rating analysis |
| • |
Charge controller analysis |
| • |
Inverter analysis |
|
| Fig. 2: |
Sizing analysis steps taken to calculate size of each component
used in a hybrid solar and wind generation system (Mousazadeh
et al., 2009) |
Resource data analysis: The availability of solar and wind resource
at the location where the generator will be used need to be obtained for 7 days
on hourly basis to obtain the weekly average value. The analysis performed will
calculate average wind speed, global insolation, direct insolation and diffuse
insolation according to the hourly global insolation data input by the user.
Total daily load demand and corrected load demand analysis: AC load
and DC load on hourly basis in a day that will be powered by the hybrid generator
are required input by the user. The total daily load demand will be calculated
according to Eq. 1. Corrected load demand is the power to
be generated by the hybrid generator to power the load taking into account the
losses from the other components used in the design. Efficiency of battery,
wiring and charge controller will be used to calculate the corrected load demand
as shown in Eq. 2 (Sandia National Laboratories,
1995):
| Where: |
| Ptotal |
= |
Total daily load demand (Wh) |
| PAC |
= |
Total AC load (Wh) |
| PDC |
= |
Total DC load (Wh) |
| ήinv |
= |
Inverter efficiency |
| Pc |
= |
Corrected load demand (Wh) |
| Pwiring |
= |
Power losses from wiring |
| Pcharger |
= |
Power losses from charge controllers |
| Pbettery |
= |
Power losses from battery |
Battery size analysis: Batteries are needed to store energy during excess
energy generation to be supplied to the load when generation is low. Battery
size depends on the Depth of Discharge (DOD) of the battery and the number of
power storage days when generation is zero. Battery voltage or the system voltage
selection depends on the total load demand of the system according to Table
1. The analysis will take into account the user input on the days of storage
needed and also battery DOD.
Deep cycle batteries can be discharged until 80% of the battery capacity but
to preserve battery life, the battery DOD is best set at 50% (Moseley,
1996; Sandia National Laboratories 1995). Equation
3 and 4 will be used to calculate battery capacity in
watt hour and ampere hour:
| Where: |
| CB(Wh) |
= |
Battery capacity (Wh) |
| CB(Ah) |
= |
Battery capacity (Ah) |
| Ptotal |
= |
Total daily power demand (Wh) |
| SD |
= |
Number of storage days |
| BDOD |
= |
Depth of discharge of battery |
| VB |
= |
Battery voltage |
Power generation ratio analysis: Power generation ratio by the solar
and wind generator is an input by the user. The ratio depends on the quantity
of available resource in the area the generator is built. If the area receives
more sunlight than wind speed, the ratio of power generated by the solar generator
will be higher than the wind generator. Equation 5 and 6
is used to calculate the total power generated by the solar based generator
and total power generated by the wind based generator:
| Where: |
| Pwind |
= |
Power generated from wind generator (Wh) |
| Psolo |
= |
Power generated from solar generator (Wh) |
| Rw |
= |
Ratio of wind generation |
| Rs |
= |
Ratio of solar generation |
| PC |
= |
Corrected load demand (Wh) |
Solar panel power rating analysis: The power rating of the solar panels
depends on the ratio of solar generation and total peak sun hours in a day.
Peak sun hour is the number of hours in a day where an average of 1000 W m-2
of solar irradiance is received. The recorded average peak sun hour for Malaysia
is around 4.37 h (Daut et al., 2011). Equation
7 is used to calculate the solar panel power rating required for the design:
| Where: |
| Prating |
= |
Solar panel power rating (W) |
| Psolor |
= |
Power generated from solar generator (Wh) |
| psh |
= |
Peak sun hour (hours) |
Wind generator rating analysis: Friction and mechanical losses in the
wind generator need to be taken into account when the generator rating is analysed.
Equation 8 and 9 is used to calculate the
wind generator power rating and swept area of the wind turbine (DNV
and Riso, 2002):
| Where: |
| Wrating |
= |
Wind generator power rating (W) |
| ηg |
= |
Wind generator efficiency |
| P |
= |
Air density (kg m-3) |
| A |
= |
Swept area (m2) |
| V |
= |
Wind speed (m sec-1) |
Charge controller analysis: Charge controllers or also known as power
conditioners are used to protect batteries from over charging, under charging
and also to maintain flow of electricity to the load. Output voltage of the
charge controller must be the same with the battery voltage and charging current
must be able to support minimal current from the generators. Equation
10 and 11 is used to calculate the charge controller
size in amps for the solar and wind generator:
| Where: |
| Cs |
= |
Solar charge controller current rating (A) |
| Cw |
= |
Wind charge controller current rating (A) |
Inverter analysis: Inverter is required to have an AC power output from
the system. Minimum AC power output from an inverter must be able to handle
surge capacity three times the maximum AC load demand. Equation
12 is used to calculate minimum inverter power:
| Where: |
| Pinv |
= |
Minimum inverter power rating (W) |
| PAC-Max |
= |
Maximum AC load power (W) |
RESULTS AND DISCUSSION
Experiment results on performance of tracking system: Experiments are
carried out to compare the performance of the solar tracker with one axis tracking
system, two-axis tracking system and a fixed solar panel as a control. Each
tracking system is installed with a 6 V, 3 W solar panels. Comparison between
the generations of power of the three systems are made by measuring the short
circuit current of the solar panels using a current sensor collected by an electronic
data acquisition system. The difference between the generations of the systems
are needed as in the sizing software developed, there are options for solar
tracking system to be selected by the user. The experiment to compare the performance
of one-axis tracking system, two-axis tracking system and a fixed solar panel
as the experiment control was done on a typical cloudy day in Universiti Teknologi
PETRONAS (UTP). The experiment was done from 8.00 a.m. till 6.00 p.m. Figure
3 shows the short circuit current generated by the fixed solar panel which
acts as the control of the experiment, one-axis tracker and two-axis tracker.
The control in the experiment starts to generate about 0.025 A of current at
the moment the experiment was started at 8.00 a.m. whereas both the solar trackers
start at current generation of about 0.05 A. It can be observed that the current
for all three systems gradually rises until about 10.00 a.m. and decreases gradually
after that period. The highest value of current generated by the fixed solar
panel at that period is about 0.052 A whereas the highest value of current generated
by the one-axis tracker and two-axis tracker are 0.13 and 0.21 A, respectively.
Current generation reduces gradually at that period due to decrease in the solar
irradiance at the earth’s surface due to passing clouds. The peak current
for all the systems at the day was at about 12.00 p.m.
|
| Fig. 3: |
Short circuit current generated by the 3W, 6V solar panels
attached to the two-axis tracker, one-axis tracker and fixed solar panel |
The control of the experiment peaks at about 0.072 A whereas the one-axis tracker
and two-axis trackers peaks at about 0.21 and 0.32 A. The current generated
for all the systems at this period is the highest because the sun is at the
highest elevation in the sky and the sun rays have the shortest path to reach
the earth’s surface through the atmosphere. After noon, the current generation
reduces and rises again at about 3.00 p.m.
The apparent trajectory of the sun which varies with position and time makes
sun tracking an essential method to increase power gain from solar power generating
systems. The incidence angle between the incident solar insolation and PV module
are minimized by sun tracking systems resulting in higher solar power gain (Singh
et al., 2012; Poulek and Libra, 1998) Figure
4 shows the angle of incidence, θ of solar radiation and the higher
the angle, the lower is the solar power gain. In a two-axis solar tracker, the
angle of incidence θ are maintained at an optimum perpendicular angle to
the solar radiation at all times which maximizes the power gain (Mousazadeh
et al., 2009). The two-axis tracker has a power gain of about 18
to 25% higher compared to the one-axis tracker. The location of Malaysia at
the equator at about 4°N makes a two-axis tracking system not suitable for
large scale electricity generation and the best would be a one-axis tracking
system.
Sizing software descriptions: The sizing software is developed using
Visual Basic 2010 Software Development Tool. The software developed needs the
user to input the following data to perform the sizing analysis:
| • |
Sun insolation and wind speed data for 7 days (hourly) |
| • |
Load to be powered |
| • |
Generation ratio |
| • |
Battery, charge controller and inverter details |
The analysis performed by the sizing software will output the following:
| • |
Total power demand and corrected load demand |
| • |
Required PV cell specifications |
| • |
Required wind turbine specification |
| • |
Required charge controller specifications |
| • |
Required battery bank specifications |
| • |
Required inverter specifications |
Figure 5 shows the main input window for the sizing analysis
software where the resource type are selected and system details are input by
the user and Fig. 6 shows the final output window where system
details will be shown after the sizing analysis have been performed by the software.
Hybrid generator assembly and testing: The software developed was used
to size a hybrid solar and wind generator with a 12 V configuration to power
an AC load of 85 Wh and DC load of 45 Wh incorporated with a one axis solar
tracker. Table 2 shows the specifications of the components
used for the hybrid solar and wind generating system. The hybrid generating
system consists of a 50 W solar based generation system and a 50 W wind based
generation system.
| Table 2: |
Hybrid solar and wind generating system specification of the
system that was built and tested |
 |
|
| Fig. 5: |
Main input window for the sizing analysis software where the
user selects the resource type and the component details of the hybrid generating
system |
|
| Fig. 6: |
Final output window where the size of all the components
to be used in the hybrid generating system are shown to the user after completion
of the analysis |
| Table 3: |
Performance of the hybrid generating system at different
solar insolation and wind speed |
 |
The system built has a storage capacity of 100 Ah at 50% battery DOD for 3
days of storage, thus the system was able to sustain the load for 3 days if
there is no any sun light or wind without deep discharging the deep cycle battery.
The wind turbine used was a horizontal axis wind turbine with a swept area of
0.6 m2. The solar and wind charge controllers have current
rating of 5 A.
Table 3 shows the hybrid generating system performance at
different wind speed and solar insolation. The charge controllers holds the
battery charging voltage constant at about 13.4 A to prevent the battery from
being overcharged. The average voltage and average current generated by the
solar generation system increases with solar insolation value. The highest solar
insolation value the system was tested is at 700 W m-2. The wind
for the testing was produced from a 160 W blower. The wind generator was tested
at maximum wind speed of about 3.5 m sec-1 and minimum wind speed
at about 2.5 m sec-1. Average voltage and average current generated
by the wind generator increases with wind speed.
Sizing analysis for the hybrid solar and wind generator are performed using
resource data collected on hourly basis for 7 days to obtain the weekly average
value in the location where the generator will be developed. Hourly solar radiation
data and wind speed data used in sizing analysis will produce the most optimum
size of components to be used in the design. An optimally sized solar and wind
hybrid generating system would be the best generation system in Malaysia as
generating systems with single energy source has disadvantages. Sunlight is
only available in the day and average wind speed in Malaysia which is at about
3-5 m sec-1 are not sufficient for a wind generation system alone
(Shafie et al., 2011; Hasan
et al., 2011).
CONCLUSION
Proper sizing analysis can eliminate the intermittent nature of wind and solar
resource. The hybrid generating system sizing software development explained
in this study will be able to make solar and wind electricity more sustainable.
The software developed for the sizing analysis uses an iterative method where
daily average sun insolation and wind speed on hourly basis makes a reliable
sizing calculation for the components to be used in the hybrid generation system.
The hybrid solar and wind generating system developed is able to sustain an
AC load of 85 Wh and DC load of 45 Wh with 3 days of storage and 50% battery
depth of discharge. The recommended future work for the software developed would
be incorporating the sizing analysis software together with a real time hybrid
PV and wind generation system performance monitoring system. The user interactive
system will be able to assist home users and industry based users to set up
and monitor their hybrid PV and wind generation system.
ACKNOWLEDGMENT
The authors are grateful to Universiti Teknologi PETRONAS for the support given
in completing this research project.
Subscribe Today
