Wind Potential Assessment Analysis of Jhampir, District Thatta Sindh, Pakistan

Wind potential analysis is analyzing how much wind energy is available in particular region. It is most important step because the economics of project depends on the site wind resources. Wind plant depends on the variation of long-term mean wind speed and other characteristics which vary from a distance to distance. This study discusses the wind speed characteristics and wind potential analysis using three years 2014-2016 wind data of Jhampir located in district Thatta, situated in the Southeast of Sindh province. The numerical Weibull distribution approach is used to estimate parameters. The correct estimation of wind parameters and class is essential before developing any wind project in the region. The data used in this study is measured at 80 m height. The region is classified as from class 1-7. The results show that monthly mean speed values lie between 4.79-10.96 m/s. The annual mean scale and shape parameters lie in the range of 7.42-7.59. The wind power density was found in a range of 303.31355.64. This study is related to the decision-making process on a significant wind project in Thatta or nearby region. The stable wind energy pattern is observed in the region for harnessing wind energy almost throughout the year. The Weibull probability density curves also indicate a trend of a boost in the chances of observing wind from 2014-2016.


INTRODUCTION
P akistan is currently facing severe energy crises due to distribution losses, Mismanagement of resources and so on. Most of the power produced in Pakistan is from fossil fuel or hydel sector, only 3% share is from the renewable energy source [1].
Depletion of natural resources like fossil fuel and climate change are serious issues that have major impacts on the economic and social development of countries.
Renewable energy sources like wind can provide an opportunity to promote sustainable development and therefore reduce dependency on fossil fuel-based energy [2]. Renewable energy share has increased around the globe during last 4 or 5 years due to overall cost reduction of the wind and solar equipment that has ultimately put the wind and solar in a competitive position to other more economic sources available in the market [3].
The aim of this paper is to estimate wind potential of Region Nonetheless, wind turbine installation in each site is viable only after accurate analysis of site feasibility. It is necessary to conduct a feasibility study for installing turbines before any major decision by authorities [4].
Jhampir to add more wind turbines in the region, located in the southwest of Sindh province. For that, Weibull distribution is used to analyze characteristics of wind using the wind data for a period of 2014, 2015 and 2016 at height of 10m. Weibull distribution is chosen because it is a most widely used method for wind data analysis.
Other methods such as Rayleigh distribution is also used but it is single parameters function and Weibull gives a better fit than Rayleigh distribution. This data may be extrapolated to 20 and 30m height for lower altitude velocity. The yearly mean wind velocity for Jhampir is 7.5 m/s which is suitable for wind turbines because wind turbine usually operates between 6-25 m/s. The extrapolation of data will help in analyzing the data at higher altitude and we know that as we go higher velocity and wind density tends to increase [5]. Few wind farms have been installed in the region selected for this study but still, this areas requires more study of wind potential in order to the accurate sitting of future turbines. Further, more accurate sitting of turbines helps in reducing negative environmental impacts of wind turbines with proper wind data characterization.

Wind Data
Data was obtained from Jhampir wind power plant located

STUDY REGION GEOGRAPHY
Sindh is one of four provinces of Pakistan, and it has more than 500-kilometer coastal range and huge wind corridor. The Thatta city is located in the southwest of Sindh province Fig. 1

Weibull Distribution Function
Wind velocity distribution study is a critical aspect to evaluate wind energy potential for a location [8]. In Weibull function, the discrepancy in the wind velocity is characterized by two functions which are PDF (Probability Density Function) and CDF (Cumulative Distribution Function) respectively [8][9]. This is a two- Where f(v) is the probability of occurrence of gust,v is velocity, k is dimensionless shape factor which shows wind velocity stability and is related to the variance of wind speed. The Shape parameter is also known as Weibull slope. The higher the value of k higher would be the stability of wind speed at given location. Where c is scale parameter (with units of wind velocity) and it is related to mean wind speed. Higher the value of c indicated that wind speed for that month is higher [9][10].
These Weibull parameters can be found using several methods, in this study the two methods Empirical and Energy pattern factor method is chosen [11]. Both of these to calculate scale and shape parameters which is a requirement for probability density function of Weibull distribution. Both of these methods are more simple and elegant compared to others methods [11][12].

Empirical Method
The empirical approach is regarded as a special case of moment method, where Weibull parameters k and c are given as [12]:

Energy Pattern Factor Method
Energy pattern factor approach is the newly developed process by Ali Akdag (2009) and it has much simpler formulation than another method to calculate Weibull parameters. Equations of Energy pattern factor method are given as [13]: Standard deviation and power density can be described mathematically as:

SELECTION OF WIND TURBINE
Even if we know that particular site has good wind resource, the selection of wind turbine is still challenging task. Wind turbines are designed according to windresource and environmental conditions. In low windareas wind turbineshave large rotors to capture more energy and in high wind regimes turbines have a smaller rotor but high rating generators [14].
Once the most apparent and maximum wind velocities are established, the wind turbine operating range can be predicted and is given by [15]: Where v co is wind velocity at which the wind turbine shuts down (cut-out wind velocity) v ci is the wind velocity at which wind turbine starts to produce power known as cut in wind velocity and v rated is the wind velocity at which wind turbine runs at full rating.The power output potential curve of a wind turbine differ from one turbine to other [16][17] (Table 2).

RESULTS AND DISCUSSION
In

TABLE 4. SCALE AND SHAPE PARAMETERS OF THE WEIBULL DISTRIBUTION
The Fig. 2 clearly indicates that the monthly mean wind speed is higher during the months of May, June, July, and August and lower during the months November, December, and January. Fig. 3(a-c) demonstrates the Weibull distribution and wind speed model justification. It clearly indicates that wind speed values are quite close and fit well with Weibull distribution. Fig. 3(a-c) also indicates data validation for chosen method. The more align the data with a straight line more suitable is a method to analyze given data.  Fig. 4(a). However, it is higher for the year 2016 in Fig.   4(b). Hence it is concluded that both methods show a significant increase in chances of wind for given region.
In probability density curves both Empirical and Epf method show, 7-8 m/s having a maximum frequency which is more than required average wind speed for any site selection.   The average of annual mean wind speed value is 7.50 Which indicates that Jhampir is a quite good location for Generating Electricity from the wind and it lies in class 4.
It is also perceived that trend in annual mean wind velocity is linear and Annual power density values of both methods are also close to each other.
Therefore a wind energy conversion system can run at theoptimum output only if it is devised for the region where it is to be applied, as rated power, and cut in, rated and cut off wind speeds must be characterized according to region [19][20].

CONCLUSION
In this study, three years of wind data (2014-2016) have been analyzed. The Weibull function was adopted to interpret the data. The monthly and annual mean wind velocity revealed a good stable wind pattern. The mean annual mean wind speed was 7.5 which ultimately puts