Investigation on available wind energy at Tungku beach

Abstract

In this paper, wind velocities and directions (sea and land) are recorded in different days and times. The data collected were compared with the weather data from the Brunei Darussalam Meteorological Service (BDMS) and the findings of other researchers and were found to be in good agreement. The potential of wind energy is predicted from the available data collected. The average generated power (forenoon and afternoon) is found to be 25 (mean) and 18 W (median), 101 (mean) and 73 W (median), 912 (mean) and 660 W (median), 10137 (mean) and 7331 W (median) for a rotor with a diameter of 2.5, 5, 15 and 50 m, respectively. The power density P_d for wind farming is found to be 0.26 (mean) and 0.19 (median), 0.31 (mean) and 0.22 (median) for the rotor whose diameter is 2.5 and 50 m, respectively, while the average P_d values are found to be 0.28 (mean) and 0.2 (median) for the rotor whose diameter is 5 and 15 m.

Keywords

wind velocity / temperature / tower height / power density / Raleigh distribution / electrical power

Introduction

The production of electrical power from renewable sources is expected to grow rapidly over the next few decades. The renewable energy, especially wind energy, is becoming more popular with each passing day. The wind turbine drives the generator to generate electrical power from the available wind. In this area, many researchers [1-4] are actively doing research to improve the utilization of wind energy for generating power. It has been found that the availability of power in a wind turbine is dependent on the wind speed, power curve of the turbine and wind fluctuation of wind energy [1]. Spancer<FootNote>

Spancer J. Wind energy production efficiency: An empirical research. Wind Power Production Inventive, Ottowa, Canada, 2005, 1-22

</FootNote> investigated on the economic aspect of wind power generation. It was found that the kilowatt per hour is dependent on wind velocity, temperature, barometric pressure and direction of wind. Guerri et al. [2] conducted a numerical simulation on the fluid flow around a roof mounted wind turbine. They computed the performance of the above type of turbine assuming the flow to be fully turbulent. They also verified their results with published data. Alaydi [3] collected wind data at two sites in Gaza Strip, namely, Gaza city and Gaza International Air Port in Rafah for assessment of wind energy potential at these two places. He recommended the sites using the data collected and assuming a tower height of 50 m. The performance evaluation and accuracy enhancement of a novel day-ahead wind power forecasting system was conducted in China [4]. This system consisted of a numerical weather prediction (NWP) model and artificial neural networks (ANNs). The NWP model was established by coupling the global forecasting system (GFS) with the weather research and forecasting (WRF) system together to predict meteorological parameters.

Improved techniques were developed to calculate the wind power resource of an offshore area [5]. This method used publicly available oceanic, environmental and socio-economic data to identify areas less suitable for development due to physical or technical constraints, safety or other hazards, environmental concerns, or competing uses. In addition, annual energy output was calculated for a representative offshore wind turbine using wind speed data from meteorological buoys. A generic power-law relationship between global warming and the usable wind energy (Betz’s law) was proposed [6]. The power law index (<4, region dependent) was then determined using simulated atmospheric parameters from eight global coupled ocean-atmosphere climate models (CGCMs). It was found that the power-law relationship held across all eight climate models and also was time scale independent. Al-Badi [7] studied the wind power cost per kilowatt-hour of energy produced using four types of wind machines at 27 locations within Oman. From the data obtained between 2000 and 2009, wind duration curves were developed and utilized to calculate the cost per kilowatt-hour of energy generated from four chosen wind machines.

From the previous research, it is seen that harnessing wind energy is dependent on available wind, which in turn is dependent on geographical location. Therefore, there is a possibility of generating power from available wind data collected. In this paper, the average generated power in the forenoon and afternoon is calculated with different rotor diameters. In addition, the mean and median power densities are calculated for different rotor diameters.

Geographical location of Tungku beach

Tungku beach is located at latitude 4.9β( 4°54′0″N) and longitude 114.8 (114°47′60″E) at an altitude of 76 m in Brunei Darussalam. It is a part of Southeastern Asia, bordering the South China Sea and Malaysia. It is slightly smaller than Delaware, U.S., with an area of 5770 km² and a coastline of 161 km.

For the above location the weather forecast data for a typical day from Thursday to Sunday is that the wind velocity is from a minimum of 0.5 m/s (1 knot) to a maximum of 4.5 m/s (9 knots). The data collected at the spot falls within the range of the above forecast data. The data collected is thought to be more applicable for computation, and therefore, they are used for analysis and computation in the paper.

Data collection and analysis

As mentioned that the weather data were collected by using the Kestrel 3000 pocket weather meter in terms of wind velocity, wind direction (sea and land), relative humidity, and temperature at Tungku beach two days in September and October 2010. The data were taken at an interval of 10 minutes for one hour for each of the month. In September, the data were collected before noon and in October they were collected in the afternoon. Each observation of v_{\mathrm{r}\mathrm{e}\mathrm{f}} (Table 1) is the average reading of 10 minutes of continuous monitoring of the wind velocities. As Brunei Darussalam is located near the equator, the temperature and humidity remains fairly same throughout the year. As such the atmospheric pressure remains more or less the same (except for adverse weather conditions) throughout the year. The atmospheric pressure was recorded by an aneroid barometer in the thermodynamics lab of Institute Technology Brunei which is close to the experimental site. The recorded wind data are mentioned in Tables 1 and 2. The measured wind velocities at a reference height of 3 m are converted to the corrected velocities at 50 m height using the following standard formula:

(1)v=v_{\mathrm{r}\mathrm{e}\mathrm{f}}{\left({\displaystyle \frac{H}{H_{\mathrm{r}\mathrm{e}\mathrm{f}}}}\right)}^{\mathrm{0.142}}.

Raleigh distribution is widely used for analyzing wind velocities. Therefore, it is decided to use this for computational purpose. The probability function of this distribution is given as

(2)f(X)={\displaystyle \frac{\mathrm{2}X}{S^{\mathrm{2}}}}{\mathrm{e}}^{-\frac{X^{\mathrm{2}}}{S^{\mathrm{2}}}},X\ge \mathrm{0.}

The expression of mean is

(3)v_{\mathrm{a}\mathrm{v}}={\displaystyle \frac{\sqrt{\pi }}{\mathrm{2}}}S.

The expression of median is given as

(4)v_{\mathrm{m}\mathrm{e}\mathrm{d}}=\sqrt{{\displaystyle \frac{\ln \mathrm{4}}{\mathrm{2}}}}S.

Using the perfect gas equation, the air density is given as

(5)\rho ={\displaystyle \frac{P_{\mathrm{a}\mathrm{t}\mathrm{m}}}{\mathrm{0.287}(t+\mathrm{273})}}.

The value of the v is calculated at a height of 50 m from Eq. (1). The v_{\mathrm{a}\mathrm{v}} was calculated from all the v values. Using Eq. (3), the S was found out. Then using S and from Eq. (4), the v_{\mathrm{m}\mathrm{e}\mathrm{d}} was calculated. Using all the collected t values (Tables 1 and 2), the t_{\mathrm{a}\mathrm{v}} was computed. Using t_{\mathrm{a}\mathrm{v}} and an average atmospheric pressure of 100 kPa, the density, \rho is calculated from Eq. (5). The above results are displayed in Tables 3 and 4, respectively.

In Brunei Darussalam, the weather station is located near the airport which is inland and a short distance from Tungku Beach. The annual average wind velocities [8] for the year 2000 to 2005 are listed in Table 5.

The variation of inland and onshore wind speed is specific to the locations. It has been found that the measured waterfront wind speeds onshore in residential areas of Victoria, British Columbia, Canada were generally about double the inland value [6].

Tables 4 and 5 (computed results at actual location (CRA)) are compared with Table 6 (weather data near airport (WDA)) and it is found that v_{\mathrm{a}\mathrm{v}}(\mathrm{C}\mathrm{D}\mathrm{A})=(\mathrm{1.3}\sim \mathrm{1.4})v_{\mathrm{a}\mathrm{v}}(\mathrm{W}\mathrm{D}\mathrm{A}). This relation is thought to be more logical as V_{\mathrm{a}\mathrm{v}}(\mathrm{C}\mathrm{D}\mathrm{A}) is measured onshore and v_{\mathrm{a}\mathrm{v}}(\mathrm{W}\mathrm{D}\mathrm{A}) is measured inland. The obtained relation is also supported by the already mentioned research [9]. The other data at the two locations also follow the same trend (The values at sea shore are higher than those inland).

From the above discussion it is observed that the v_{\mathrm{a}\mathrm{v}} (Tables 3 and 4) which is approximately 3 to 4 m/s, and calculated at the investigated site is quite reasonable. As such there is a high probability that a velocity of 3-4 m/s will be available throughout the year. Since a wind turbine is possible if the wind speed is more than 3 m/s, it is recommended wind turbines be installed at a height of 50 m at this location. The power is calculated using Eq. (6).

(6)P_n={\displaystyle \frac{C_{\mathrm{p}}{\mu }_{\mathrm{c}}A\rho v_n^{\mathrm{3}}}{\mathrm{2}}},

where n=\mathrm{a}\mathrm{v}\mathrm{o}\mathrm{r}\mathrm{m}\mathrm{e}\mathrm{d}.

Using C_{\mathrm{p}}=\mathrm{0.386}, {\mu }_{\mathrm{c}}=\mathrm{0.5} and Eq. (6), the mean and median power for a single turbine of three different sizes, namely small (), medium () and large (d>\mathrm{50}\mathrm{m}), have been computed. Assuming a lateral spacing of \mathrm{3}d and a longitudinal spacing of \mathrm{10}d between the towers as illustrated in Fig. 1, the power generated for two rows of turbine has been calculated. The power density for the three types of turbine has also been computed. The results have been mentioned in Tables 5 and 6.

Results and discussions

From Tables 1 and 2, it is noticed that the wind direction changes from sea to land between forenoon and afternoon. This is a natural phenomenon and the reasons for such are found in any standard textbook on meteorology, as such it will not be discussed here. From the above tables it is also observed that in general the wind velocities are less in the forenoon than afternoon. As such it is expected from the present investigation that the power generation in the forenoon will be less than that in the afternoon.

It is known that a wind turbine can work effectively in the range of . When , the turbine may not rotate, as it has insufficient energy to overcome the friction. When v>\mathrm{25}\mathrm{m}/\mathrm{s}, the turbine rotor rotates at a very high speed and there is a possibility of damage. After computation from Tables 3 and 4, the following values of wind velocities of v_{\mathrm{a}\mathrm{v}}=3.17 m/s and 3.92 m/s, v_{\mathrm{m}\mathrm{e}\mathrm{d}} = 2.98 m/s and 3.69 m/s are found for forenoon and afternoon, respectively. As the computed wind velocity is within the range of , it is expected that the installed wind turbine will work properly. Since, Brunei Darussalam is located near the equator; as such there is very little variation in temperature during the day, which can be observed from Tables 3 and 4. As such there is a little change in the density of wind (\rho=1.16 kg/m³ and 1.15 kg/m³, at forenoon and afternoon respectively). From Eq. (5), it is found that p\propto \rho v^{\mathrm{3}} and it is also observed (as mentioned in Tables 3 and 4) that there is little change in the values of \rho and v throughout the present investigation. Therefore, it is believed that there will be a more or less steady extraction of wind energy from any installed turbine in this location.

The values of P using three types of rotor and the corresponding P_{\mathrm{d}} values using V_{\mathrm{a}\mathrm{v}} and V_{\mathrm{m}\mathrm{e}\mathrm{d}} is presented in Tables 5 and 6. In general, it has been found that values P and P_{\mathrm{d}} is always less in the forenoon than in the afternoon in this present investigation. The average generated power (forenoon and afternoon) is found to be 25 (mean) and 18 W (median), 101 (mean) and 73 W (median), 912 (mean) and 660 W (median), 10137 (mean) and 7331 W (median) for a rotor with a diameter of 2.5, 5, 15 and 50 m, respectively. From the above results it can be seen that the turbine with a diameter of 50 m might be suitable for small household living near this area. Turbines with different sizes might be used for other small scale power requirement.

From Tables 5 and 6, the average (forenoon and afternoon) values of P_{\mathrm{d}} is found to be 0.26 (mean) and 0.19 (median), 0.31(mean) and 0.22 (median) for a turbine with a diameter is 2.5 and 50 m, respectively. For a turbine whose diameters is 5 and 15 m, the average P_{\mathrm{d}} values are 0.28 (mean) and 0.2 (median), respectively. Form the above discussions it can be stated that if it is decided to use the present location to utilize wind energy for power farming, then the turbine with diameters of 5, 15 or 50 m might be suitable depending on cost analysis. In the present investigation, the P_{\mathrm{d}} values are low (0.7 W/m²) compared to other [8]. All the power calculations are based on assuming the values of coefficient of performance (0.275) and conversion efficiency (0.7). The P and P_{\mathrm{d}} values can be improved by improving on C_{\mathrm{p}} and {\mu }_{\mathrm{c}}. The value of C_{\mathrm{p}} can be improved by selecting a suitable turbine. On the other hand {\mu }_{\mathrm{c}} is comprised of two components, namely the transmission efficiency and the generator itself. To improve the transmission efficiency, it is suggested that mechanical components and lubrication be looked into. More power can be generated by selecting generators with minimum losses.

Conclusions

Wind velocities are collected in Tungku Beach in the forenoon and afternoon. These wind data were compared with weather data from BDMS. The mean and median of wind velocities are calculated using Raleigh distribution at a reference height of 3 m. It is found that the calculated average velocities are in good agreement with the findings of other researchers. From the calculated wind velocities, the velocities at a height of 50 m are estimated using appropriate formula.

The power density P_{\mathrm{d}} for wind farming are found to be 0.26 (mean) and 0.19 (median), 0.31 (mean) and 0.22 (median) for a rotor with a diameter of 2.5 and 50 m, respectively. For the rotor with a diameter of 5 and 15 m, the average P_{\mathrm{d}} are found to be 0.28 (mean) and 0.2 (median) respectively. Based on the obtained results, it is seen that the rotor with a diameter of 50 m has both the highest mean and median power density values. Therefore, it is recommended the rotor with a diameter of 50 m be used for power generation.

Notations

References

[1] Boukhezzar B, Siguerdidjane H, Hand M M. Non-linear control of variable speed wind turbines for generator torque limiting and power optimization. ASME Transactions on Journal of Solar Energy Engineering, 2006, 128(4): 516–531

[2] Guerri O, Sakout A, Hamdouni A. Numerical simulation of the fluid flow around a roof mounted wind turbine. Journal of Wind Engineering, 2010, 34(5): 501–515

[3] Alaydi J Y. A survey on the assessment of wind energy potential in Gaza Strip. Journal of Wind Engineering, 2010, 34(5): 531–541

[4] Zhao P, Wang J F, Xia J R, Dai Y P, Sheng Y X, Yue J. Performance evaluation and accuracy enhancement of a day-ahead wind power forecasting system in China. An International Journal of Renewable Energy, 2012, 43: 234–241

[5] Sheridan B, Baker S D, Pearre N S, Firestone J, Kempton W. Calculating the offshore wind power resource: Robust assessment methods applied to the U.S. Atlantic Coast. An International Journal of Renewable Energy, 2012, 43: 224–233

[6] Ren D D. Effects of global warming on wind energy availability. Journal of Renewable and Sustainable Energy, 2010, 2(5): 1–5

[7] Al-Badi A H. Wind power potential in Oman. International Journal of Sustainable Energy, 2011, 30(2): 110–118

[8] Newsletter on Energy and Technology, The Energy Advocate, P.O. Box 7609, Pueblo West, CO 81007–0609, USA

[9] Tuller S E. Onshore flow in an urban area: Microclimatic effects. International Journal of Climatology, 1995, 15(12): 1387–1398

RIGHTS & PERMISSIONS

Higher Education Press and Springer-Verlag Berlin Heidelberg