Therefore, average time required for battery charging = [(27 � 157)/60]h = 70.65 h.

Experimental observation:

It takes 8.1 h to charge 1 V of the battery [from the previous result section]. Moreover, after 5 V, it takes 18.8 h to charge another 0.5 V.

Therefore, total estimated battery charging hour = [(8.1 � 5) + (18.8 � 2)]h = 78.1 h.

Percentage of error:

$$\text{Percentage of Error, POE} = \frac{(|70.65 - 78.1|)}{70.65} \times \mathbf{100\%} = \mathbf{10.54\%} \tag{3}$$

would result in damaging the battery. As far as wind speed of 4 m/s was concerned, the energy harvesting circuit, taking only 10.4 h to charge up the battery, again showed an excellent performance of 31% efficiency comparing with direct charging that took a straight 15 h lap. For 3 m/s, the energy harvesting circuit still held the top position

Supercapacitor-Based Hybrid Energy Harvesting for Low-Voltage System

http://dx.doi.org/10.5772/intechopen.71565

19

To recapitulate, this research provides an excellent novel idea of a stand-alone Maglevbased VAWT system connected to a PMSG that can harvest energy via Supercap-based battery charging circuit in low wind areas of rural areas. Research contribution is original, and it gives an outstanding foundation for future study in energy harvesting for low wind

The entire research had not been absolutely smooth all throughout and naturally it faced few

i. Firstly, turbine blade design was not taken into consideration in the simulation. As there was no proper mathematical model that relates turbine blade number to output torque or power, the simulation therefore did not account for blade design although it could give better performance if blade number was included in the design. It was not possible to apply finite element analysis (FEA) on turbine blades due to lack of time. Moreover, the position of blade, cut-in angle and vibration analysis of the turbine could be done with FEA. Surely it could have given a wider research scope area on modeling, and blade material could have been brought into the optimization

ii. Moreover, DC/DC boost converter used in this research did not perform well according to the data sheet in its minimum range. As it was stated in the data sheet, the converter can step up voltage from as low as 2.5 V, practically it could not step up voltage less than 4 V. Therefore, the Supercap charging range was made from 4 to 7.5 V, which should have

Conventional DC/DC boost converter is to be replaced with the efficient one, which is specifically designed to work with voltage as low as 2–3 V. This will help Supercap to discharge even more and will play a vital role while dealing with low wind. All these changes will improve the system and should make it capable of performing at 2 m/s. Since most of the electronic devices operate at 12 V, a second DC/DC converter may be placed to charge a 12 V battery

Laptop should be replaced with wireless system in the future. A real-time wireless monitoring interface could be made available. Embedded solutions providing wireless end point connec-

handsomely with 28% efficiency in comparison with direct charging.

The limitations of the developed system and technique are listed below:

been 3–7.5 V. This had a direct effect on system efficiency.

tivity to devices like XBEE modules can be of use in cases like this.

process for a better configuration.

rural areas.

ups and downs.

5. Future work

from the current 6 V-led acid battery.

The voltage drops in boost converter and MOSFET switch are the main reasons for difference in theoretical and experimental values.

Direct charging without converter:

The battery takes 15 h to charge 1 V from the turbine until 5 V. After 5 V, it takes 24.2 h to charge 0.5 V.

Therefore, average time required for battery charging = [(15 � 5) + (24.2 � 2)]h = 123 h

Efficiency of Supercap � based Battery Charging Circuit <sup>¼</sup> ð Þ <sup>j</sup><sup>123</sup> � <sup>78</sup>:1<sup>j</sup> 123 � 100% ¼ 36% (4)

#### 4. Conclusion

As a conclusion to this research, the achievements are reviewed in terms of research objectives. This consequently facilitates the system, and results are to be analyzed in terms of the percentage and degree of the research objectives that were achieved. Three cases had been compared for performance analysis. "Case A" showed a battery of 6 V, 3.2 AH, being charged from 4.2 to 5 V through a DC/DC converter followed by a series of four Supercaps. "Case B" and "Case C" demonstrated the direct charging of the battery, where "Case B" was experimented with the converter and "Case C" was without converter. Investigation was carried for 3, 4 and 5 m/s wind speed. "Case C" was taken as a reference. For a wind speed of 5 m/s, the result showed an increase of 19% of the charging time for Case A while charging through the Supercap. It took only 8.1 h whereas direct charging without converter took 10 h. Supercap-based charging was also found to be 133% more efficient than direct battery charging with a converter. Keeping in mind, direct charging might not be the appropriate way of charging a device since fluctuation of wind would result in damaging the battery. As far as wind speed of 4 m/s was concerned, the energy harvesting circuit, taking only 10.4 h to charge up the battery, again showed an excellent performance of 31% efficiency comparing with direct charging that took a straight 15 h lap. For 3 m/s, the energy harvesting circuit still held the top position handsomely with 28% efficiency in comparison with direct charging.

To recapitulate, this research provides an excellent novel idea of a stand-alone Maglevbased VAWT system connected to a PMSG that can harvest energy via Supercap-based battery charging circuit in low wind areas of rural areas. Research contribution is original, and it gives an outstanding foundation for future study in energy harvesting for low wind rural areas.

The entire research had not been absolutely smooth all throughout and naturally it faced few ups and downs.

The limitations of the developed system and technique are listed below:

