The increasing use of fossil fuels has a significant impact on the environment and ecosystem, which increases the rate of pollution. Given the high potential of renewable energy sources in Yemen and other Arabic countries, and the absence of similar studies in the region. This study aims to examine the potential of wind energy in Mokha region. This was done by analyzing and evaluating wind properties, determining available energy density, calculating wind energy extracted at different altitudes, and then computing the capacity factor for a few wind turbines and determining the best. Weibull speed was verified as the closest to the average actual wind speed using the cube root, as this was verified using 3 criteria for performance analysis methods (

The increase in population and rate of industrialization has led to a rise in energy demand. Fossil fuels cannot meet this demand because they negatively affect the environment and ecosystem, causing a significant increase in pollution. In other words, the energy industry and the environment are in significant crises today. Today in modern societies, energy is the most important indicator of economic growth and many countries worldwide are taking steps toward achieving a renewable energy model to solve this crisis [

As seen in literature [^{2}, and as of 2016, the population was about 26,687,000 people. Yemen has a high potential for renewable energy sources such as solar, wind, and geothermal energy [

In 2009, the Yemeni government approved the National Renewable Energy and Efficiency Strategy, which aims to increase 15% of energy efficiency (EE) in the energy sector by 2025, and target renewable energy (RE) capacity (Geothermal energy 160 megawatts, concentrated solar power 100 megawatts, solid biomass 6 megawatts, solar photovoltaic system 8.25, and wind power 400 megawatts) of total electricity by 2025. The Yemeni energy sector consists of oil, natural gas, and biofuel production. Energy production in 2012 was “15.109 kilotons of oil equivalent (ktoe), while consumption was 6,923 kilotons” [

The objective of this study is divided into three parts are: to find a probability distribution function that gives the best estimate of annual wind energy and is close to the actual value of this energy. Using the wind rose chart, the prevailing wind direction during the seasons of the year is determined at the study site. Study the characteristics of the three classes of wind turbines, calculate the capacitance factor for each of them, and determine the most appropriate class for the study site.

The rest of the paper has four more sections. Section 2 provides the methodology. Section 3 provides the basic calculations of the proposed system model. Section 4 presents the results and discussion, and Section 5 offers conclusions.

The research consists of two main parts, and each section consists of its own parts. In the first section, the statistical analysis of the measured wind speed data from the OYMK monitoring station was performed at a height of 3 m in the MOKHA area for five years (2011 to 2015), and these speeds will be converted to be suitable for wind turbine heights of 30 and 100 m.

Since the statistical values of wind speed measurements do not express the power that can be derived from these measured speeds, so the probability distribution functions that help us in evaluating wind energy from the statistical values of wind speed will be used only (without knowing the measured values).

To demonstrate the accuracy of the work of these functions, the cubic statistical values of wind speeds will be calculated and the actual power values compared to the power evaluated by the probability distribution functions, where the most used functions for the probability distribution of the wind speed measured at a given location are the Weibull and Rayleigh distributions.

In the second section, power factors for the three popular classes of rated wind turbines according to the International Electrotechnical Commission (IEC) were calculated, and the wind rose analysis was made in order to determine the most appropriate direction for steering the turbine. Also, in this section, the amount of energy that can be harvested during the year with one turbine with a rated capacity of 3450 kilowatts will be calculated, and the amount of energy required to meet the needs of Mokha region will be determined.

The Mokha region is located at longitude 43°16′60′ east and latitude 13°19′0′ north. It overlooks the Red Sea in southwestern Yemen and is 75 km north of Bab al-Mandab, 100 km west. The city of Taiz, and 170 km south of Hodeidah. The OYMK monitoring station is located 8 km south of Mokha, at a longitude 43°16′60″ east and 13°15′00″ north latitude, at a height of 3 m from the ground. The location of the OYMK monitoring station and the distance between it and Mocha.

Observation | Value |
---|---|

Air temperature [F] | 80.6 |

Maximum air temperature [F] | 80.6 |

Minimum air temperature [F] | 77 |

Dew point [F] | 69.8 |

Wind direction | S (180 degrees) |

Wind speed [m/s] at h = 10 m | 9.3 m/s |

Year | 2011 | 2012 | 2013 | 2014 | 2015 | Yearly | Sector 30° direction} | ||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|

Month | |||||||||||||

Jan | – | – | 4.73 | 2.21 | 5.65 | 1.99 | 6.12 | 1.53 | 4.32 | 2.43 | 5.17 | 2.20 | S |

Feb | – | – | 6.01 | 2.22 | 6.03 | 2.02 | 5.23 | 2.17 | 7.68 | 3.88 | 6.24 | 2.83 | S |

Mar | – | – | 5.03 | 2.10 | 5.51 | 1.87 | 6.90 | 2.46 | 7.51 | 0.89 | 6.05 | 2.23 | S |

Apr | – | – | 5.10 | 2.02 | 3.87 | 1.44 | 5.84 | 1.97 | – | – | 4.87 | 1.99 | S |

May | – | – | 4.15 | 1.54 | 3.53 | 1.57 | 3.41 | 1.64 | – | – | 3.71 | 1.62 | S |

Jun | – | – | 1.74 | 0.97 | 2.01 | 0.85 | 2.73 | 2.19 | – | – | 2.18 | 1.53 | N/NW |

Jul | – | – | 1.80 | 1.07 | 2.11 | 1.32 | 2.58 | 1.39 | – | – | 2.16 | 1.31 | N/NW |

Aug | 2.45 | 0.41 | 2.14 | 1.25 | 1.75 | 1.57 | 1.89 | 1.12 | – | – | 1.98 | 1.29 | N/NW |

Sep | 2.60 | 1.03 | 2.87 | 1.36 | 2.53 | 1.01 | 1.81 | 0.63 | – | – | 2.45 | 1.11 | N/NW |

Oct | 4.37 | 1.62 | 5.18 | 1.20 | 3.57 | 1.89 | 2.83 | 1.36 | – | – | 4.00 | 1.77 | S |

Nov | 3.99 | 1.83 | 5.44 | 1.08 | 5.99 | 1.12 | 3.82 | 0.99 | – | – | 4.85 | 1.60 | S |

Dec | 5.09 | 1.87 | 4.87 | 1.21 | 5.14 | 1.98 | 4.56 | 1.43 | – | – | 4.97 | 1.71 | S |

Annual | 3.92 | 1.84 | 4.12 | 2.16 | 3.89 | 2.22 | 3.92 | 2.37 | 6.25 | 3.29 | 4.11 | 2.35 |

In this study, environmental data available on the Iowa Environmental Mesonet (IEM) was analyzed. IEM collects environmental data from cooperating members with observing networks. The data are stored and made available on the website [

The data analyzed are wind data extracted from the integrated surface database (ISD) of the OYMK monitoring station, and the data block consisted of 16315 wind speed measurements on an hourly head (between August 23, 2011 and March 18, 2015), and the largest measured value was wind speed It is 92 m/s while the lowest value is 0.52 m/s, and the temperature ranged between 0^{○} and 380. We note also that the data for the years 2011 and 2015 are incomplete.

In order to simplify the analysis, these data were summarized by calculating the average value for each day and we reduced the data block to 1,221 average daily wind speed.

The probability density function of a Weibull distribution can be expressed as given in

The cumulative distribution function of a Weibull distribution is expressed by the integration of the probability density function as given in

As the two parameters of the Weibull distribution are shape factor

The Rayleigh distribution is a special case of a Weibull distribution with the shape parameter value of the constant

The mean value of wind speed

Sometimes two wind amplitudes should be taken seriously to assess the wind energy potential of a site, namely the most probable wind speed

The arithmetic statistics express the mean value and the standard deviation of the wind speed measured as given by

However, because the energy extracted from the wind is proportional to the cube of the wind speed and not with the speed itself, it is better to use cubic statistics. Cubic statistics express the mean value and standard deviation of a cube of wind speed measured as given by

where

In order to determine the performance of the Weibull and Rayleigh models a quantitative evaluation of the performance of these models must be made. There are different methods of evaluation are Coefficient of determination, Root mean square error (RMSE), and Coefficient of efficiency (COE).

RMSE verifies the accuracy of the model by checking the value obtained by the probability distribution function and the raw data measured. The lower the RMSE, the better the probability distribution function. RMSE ranges from

Another way to check the efficiency of a probability distribution function is to use the Coefficient of efficiency COE. The closer the value of COE to the value of the integer one, the better the accuracy of the efficiency of the probability distribution function, COE is calculated using

where

Power density assessment is of primary importance in assessing wind energy in a given region. The density of wind power depends on the density of the air

The power density can also be evaluated using the probability distribution functions as given by

Where

The observed data used in this study were measured at a height of 3 m. whereas most trade wind turbines have different hub heights ranging from 30 m to more than 100 m. Hence, Hillman’s exponential law was used to extrapolate the wind speed data to the appropriate height for our study (100 m) as shown in

where

The monthly energy index values for all sites are generally in the Mokha region less than

The capacity factor is an important parameter for selecting a suitable wind turbine for the studied site, and is defined as the ratio of the average power output

The capacitance factor can also be calculated by

where

Given the value of the capacity factor, the estimated value that a turbine with rated power

The arithmetic Statistics of the speeds measured at a height of 3 m are shown in

It is evident from

In this study, the cubic statistics of the measured wind speed were also calculated using

Statistics | Arithmetic statistics | Cubic statistics | |
---|---|---|---|

Month | |||

Jan | 5.17 | 2.20 | 5.95 |

Feb | 6.24 | 2.83 | 7.46 |

Mar | 6.05 | 2.23 | 6.77 |

Apr | 4.87 | 1.99 | 5.54 |

May | 3.71 | 1.62 | 4.29 |

Jun | 2.18 | 1.53 | 3.21 |

Jul | 2.16 | 1.31 | 2.84 |

Aug | 1.98 | 1.29 | 2.78 |

Sep | 2.45 | 1.11 | 2.92 |

Oct | 4.00 | 1.77 | 4.65 |

Nov | 4.85 | 1.60 | 5.32 |

Dec | 4.97 | 1.71 | 5.49 |

Annual | 4.11 | 2.35 | 5.19 |

Weibull and Rayleigh's parameters are computed for monthly and yearly arithmetic statistics using

Statistics | Arithmetic statistics | Weibull | Rayleigh | Cubic statistics | ||
---|---|---|---|---|---|---|

Month | ||||||

Jan | 5.17 | 2.20 | 5.82 | 2.53 | 6.57 | 5.95 |

Feb | 6.24 | 2.83 | 7.04 | 2.36 | 7.94 | 7.46 |

Mar | 6.05 | 2.23 | 6.78 | 2.96 | 7.65 | 6.77 |

Apr | 4.87 | 1.99 | 5.48 | 2.65 | 6.18 | 5.54 |

May | 3.71 | 1.62 | 4.18 | 2.47 | 4.72 | 4.29 |

Jun | 2.18 | 1.53 | 2.40 | 1.47 | 2.71 | 3.21 |

Jul | 2.16 | 1.31 | 2.43 | 1.73 | 2.74 | 2.84 |

Aug | 1.98 | 1.29 | 2.21 | 1.59 | 2.49 | 2.78 |

Sep | 2.45 | 1.11 | 2.77 | 2.36 | 3.12 | 2.92 |

Oct | 4.00 | 1.77 | 4.51 | 2.42 | 5.09 | 4.65 |

Nov | 4.85 | 1.60 | 5.41 | 3.33 | 6.10 | 5.32 |

Dec | 4.97 | 1.71 | 5.55 | 3.18 | 6.27 | 5.49 |

Annual | 4.11 | 2.35 | 4.63 | 1.86 | 5.22 | 5.27 |

The evaluation of wind resources includes the calculation of the power density generated by the wind at the study site by

The wind power density is directly proportional to the cube of the wind speed according to the relation 19, so the process of calculating the power density from arithmetic statistics for measurements of the wind speed

If

Statistics | Arithmetic statistics | Cubic statistics | Weibull | Rayleigh | ||||
---|---|---|---|---|---|---|---|---|

Month | ||||||||

Jan | 5.17 | 84.64 | 5.95 | 129.02 | 6.00 | 132.02 | 6.41 | 161.65 |

Feb | 6.24 | 148.82 | 7.46 | 254.29 | 7.37 | 245.12 | 7.74 | 284.22 |

Mar | 6.05 | 135.64 | 6.77 | 190.05 | 6.79 | 192.00 | 7.51 | 259.04 |

Apr | 4.87 | 70.74 | 5.54 | 104.14 | 5.59 | 107.17 | 6.04 | 135.11 |

May | 3.71 | 31.28 | 4.29 | 48.36 | 4.33 | 49.68 | 4.60 | 59.74 |

Jun | 2.18 | 6.35 | 3.21 | 20.26 | 3.06 | 17.59 | 2.70 | 12.12 |

Jul | 2.16 | 6.17 | 2.84 | 14.03 | 2.83 | 13.95 | 2.68 | 11.79 |

Aug | 1.98 | 4.75 | 2.78 | 13.16 | 2.69 | 11.94 | 2.46 | 9.08 |

Sep | 2.45 | 9.01 | 2.92 | 15.25 | 2.90 | 14.93 | 3.04 | 17.20 |

Oct | 4.00 | 39.20 | 4.65 | 61.58 | 4.69 | 63.29 | 4.96 | 74.87 |

Nov | 4.85 | 69.88 | 5.32 | 92.22 | 5.34 | 93.31 | 6.02 | 133.45 |

Dec | 4.97 | 75.19 | 5.49 | 101.35 | 5.51 | 102.34 | 6.17 | 143.61 |

Annual | 4.11 | 42.52 | 5.27 | 89.65 | 5.23 | 87.76 | 5.10 | 81.21 |

For further verification, the performance analysis methods are computed according to

Performance analysis method | Min value | Max. value | Optimum value | Weibull | Rayleigh |
---|---|---|---|---|---|

0 | 1 | 0 → 1 | 0.9984 | 0.9148 | |

RMSE | 0 | 1 | 0 ← 1 | 0.0632 | 0.4680 |

COE | 0 | 0 → 1 ← |
1.0280 | 1.3322 |

Since the height of the tower in the wind turbine ranges between 30 and 100 m, we should use Hellman’s exponential law (

Statistics | Arith. | Cubic | Weibull | Rayleigh | |||
---|---|---|---|---|---|---|---|

Height (m) | |||||||

3 | 4.11 | 5.27 | 89.65 | 5.23 | 87.76 | 5.1 | 81.21 |

30 | 6.37 | 8.16 | 333.1 | 8.10 | 326.1 | 7.90 | 301.7 |

50 | 7.02 | 8.99 | 445.7 | 8.93 | 436.3 | 8.70 | 403.7 |

100 | 8.01 | 10.3 | 661.6 | 10.18 | 647.7 | 9.93 | 599.3 |

Wind direction analysis is very important to calculate wind power density. In previous studies, the wind rose diagram was very popular to know the direction of the wind, and from this analysis, it is possible to know the direction in which the maximum wind energy can be generated. Returning to

Direction | Wind speed (m/s) at h = 3 m | Total | Days/year | ||||||||
---|---|---|---|---|---|---|---|---|---|---|---|

Clam | 1–3.9 (%)} | 4–5.9 (%)} | 6–7.9 (%)} | 8–9.9 (%)} | 10–11.9 (%)} | 12.0+ (%)} | Percent (%)} | ||||

345°–14° | N | 6.17% |
3.344 | 0.732 | 0.224 | 0.067 | 0.007 | 0.007 | 4.38 | 15.99 | 3.23 |

15°–44° | N/NE | 0.105 | 0.052 | 0.015 | 0.015 | 0.007 | 0 | 0.194 | 0.71 | 4.28 | |

45°–74° | NE/E | 0.015 | 0 | 0 | 0 | 0 | 0 | 0.015 | 0.05 | 2.45 | |

75°–104° | E | 0.149 | 0.075 | 0.022 | 0.022 | 0.007 | 0.03 | 0.305 | 1.11 | 4.99 | |

105°–134° | E/SE | 0.269 | 0.037 | 0.037 | 0.045 | 0.022 | 0 | 0.41 | 1.5 | 4.25 | |

135°–164° | SE/S | 3.038 | 1.232 | 0.478 | 0.164 | 0.06 | 0.007 | 4.98 | 18.17 | 3.83 | |

165°–194° | S | 10.742 | 12.586 | 18.16 | 12.56 | 6.151 | 2.187 | 62.39 | 227.7 | 5.27 | |

195°–224° | SW/S | 1.515 | 0.269 | 0.022 | 0 | 0 | 0 | 1.81 | 6.59 | 2.88 | |

225°–254° | SW/W | 0.866 | 0.067 | 0 | 0 | 0 | 0 | 0.933 | 3.41 | 2.63 | |

255°–284° | W | 1.784 | 0.127 | 0.037 | 0.007 | 0.015 | 0.007 | 1.98 | 7.22 | 2.82 | |

285°–314° | NW/W | 2.695 | 0.426 | 0.09 | 0.015 | 0 | 0 | 3.23 | 11.77 | 2.94 | |

315°–344° | NW/N | 8.853 | 3.307 | 0.911 | 0.134 | 0.007 | 0.007 | 13.22 | 48.25 | 3.46 |

Note: S = South, N = North, E = East, W = West, SE = Southeast, ES = Southwest, NW = Northwest, and NE = Northeast.

It is evident from

The average value of the speed of the south wind during this period is equal to 5.27 m/s at a height of 3 m, so the direction of the south is the direction chosen in this study.

The proposed turbine capacity is a turbine with a capacity of 3.45 Mw, as one turbine is sufficient to meet the basic needs of 1452 homes in Mokha region, and since the height of the tower is 100 m, we should use Hellman’s exponential law (

Direction | Wind speed (m/s) at h = 100 m | Total | Days/year | Vm m/s | |||||||
---|---|---|---|---|---|---|---|---|---|---|---|

Clam | 4.8 | 6.9 | 13.5 | 17.4 | 21.3 | 23.4+ | Percent | ||||

165°–194° | S | 6.17% | 10.74% | 12.59% | 18.16% | 12.56% | 6.15% | 2.19% | 62.4% | 227.7 | 10.3 |

Wind energy is harvested in the turbine between the cut-in speeds and the cut-off, which means that the efficient wind speeds that could be used to generate power fall between these two values. The analysis is based on the cut-in and cut-off speeds of the wind turbine.

The IEC has classified wind turbines into three classes.

In this study, the capacity factor was calculated using

It is noticeable that Class 3 turbines have the largest capacity parameter, which makes them the best choice, but there are other criteria that must be met to choose the turbine besides the capacity factor standard.

Generic 3.45 MW | IEC Class 1 | IEC Class 2 | IEC Class 3 |
---|---|---|---|

Rated power (kW) | 3450 | 3450 | 3450 |

Rated Power Wind speed m/s | 13 | 12 | 11.5 |

Cut-in m/s | 4 | 3 | 3 |

Cut-out m/s | 25 | 22.5 | 22.5 |

Design annual average wind speed at hub height m/s | 10 | 8.5 | 7.5 |

Rotor diameter (m) | 112 | 126 | 136 |

Power Density P_{D} (w/m^{2}) |
350.2 | 276.7 | 237.2 |

Hub heights (m) | 100 | 100 | 100 |

Mean turbulence intensity at 15 m/s I_{15} |
18% | 16% | 12% |

50-year extreme wind speed over 10 min |
50 | 42.5 | 37.5 |

50-year extreme gust over 3 s |
70 | 59.5 | 52.5 |

Capacity factor CF | 0.533 | 0.625 | 0.671 |

The other criteria was the mean turbulence intensity

It is noted the class I turbine have the top values for those three criteria, and because the study site is a site exposed to storms, so the Generic 3.45 MW-IEC Class 1 turbine is preferred for this site.

To calculate the amount of energy a class 1 turbine generates during the year if it is directed to the south,

In order to calculate the amount of energy a Class 1 turbine generates during the year, if it is directed to the south during 227 days,

Generic 3.45 MW Class I | Annual MWh | Daily KWh |
---|---|---|

Annual Energy in MOKHA MWh | 10,021 | 27,838 |

Total Energy to each household | 6.93 | 19 |

The study site is exposed to high-speed winds, especially the southern winds, which have an average speed of 10 m/s at a height of 100 m, which blow from 227 days from the day of the autumn solstice at the end of December until the summer solstice in June of the following year. It has been proved that the estimated power density by the Weibull probability distribution function