1 Introduction

Energy is the most important factor in society for sustainable development and economic growth. Today, fossil fuels are still the dominant source of energy regardless of their environmental concerns [1]. In the UN 2015 Conference for climate change, many countries chose to fight against climate change and took necessary measures. However, the withdrawal of some countries from this agreement, the U.S. for instance, has challenged the achievement of this agreement. Thus, the necessity for reducing dependency on fossil fuels is even more crucial than in the past. In this regard, many countries adopt new policies to prioritize renewable energy over fossil fuels. Therefore, the role of renewable energy in achieving sustainable development and environmental protection [2] has significantly increased. Such achievement also benefits from the reduction in cost, technology advancement, and improvement in wind farm efficiency. Currently, the technology of harvesting energy from wind is a distinguished industry in the field of green energy and as a result, wind power harvesting is competitive to the conventional sources of energy [3].

Iran is among the countries with the largest oil and gas reservoir and plays a significant role in energy organizations such as The Organization of the Petroleum Exporting Countries (OPEC). Iran heavily depends on natural gas and oil for its domestic energy demand. Such dependency is the motivation for exploring new sources of energy other than natural gas and oil. Also, fossil fuel resources are limited and the need for replacing them with renewable and sustainable energy sources is the major concern of many countries. Iran has enormous potential for harvesting wind and solar energy but pushes aside by its wealth of oil and gas. According to the Iranian Renewable Energy and Energy Efficiency Organization, solar and wind energy are the dominant utilized renewable energy in Iran with 43% and 41% market share, respectively. The total installed renewable energy power plants reach 700 MW in 2019. The geopolitical location of Iran assists the country to be also a major electricity exporter to its neighbors such as Iraq. On the other hand, energy demand in Iran grows significantly. Therefore, to meet the energy demand both for domestic use and export, it is necessary to seek other sources of energy. To assure potential investors to invest in any wind farm project, it is necessary to prepare a technical and economic feasibility study. Hence, the feasibility study provides the investors an outlook regarding the financial prospects of the project. The main contributions of the proposed framework can be summarized as follows:

  • A novel assessment of the wind energy potential in historical cities has been investigated.

  • A new evaluation method for the annual power density, energy density and capacity factor are proposed.

  • We introduce the unit cost of energy (UCE) as the economic feasibility metric indicator in Internet of Things (IoT) oriented energy harvesting and provide insight into the best wind turbine option for implementation.

  • We emphasize the importance of energy demand reduction in smart cities and shed the light on the future of wind energy planning.

2 Related works

Wind energy potential has been investigated by several researchers in order evaluate the possibility of utilizing such energy. For such investigation, it is required to predict the wind speed. There are several probability density functions such as Rayleigh function, gamma function, beta function, lognormal function, Weibull distribution function, and logistical function, which researcher previously utilized for frequency of wind speed curve [4]. Some of these functions are special cases of another distribution, for instance the chi-square and Weibull distributions are special cases of generalized gamma distribution [5].

Among these statistical models, Weibull distribution has been extensively employed for wind speed in studies [6,7,8]. Weibull function has demonstrated satisfactory result and accurately fit the wind speed frequency. Weibull distribution function can be employed as two-parameter or three-parameter distribution function. While two-parameter Weibull distribution function has convenience and flexibility relative to other statistical distributions, three-parameter Weibull distribution was suggested for representation of low wind speed [9]. Tuller and Brett [10] investigated the goodness of fit of the Weibull and Rayleigh distributions to the distributions of observed wind speeds in a topographically diverse area and concluded three-parameter has a better fit over two-parameter Weibull distributions and Rayleigh distribution for their study. On the other hand, Bowden [11] reported that three-parameter distribution offers no significant improvement over two-parameter Weibull distribution. Several researchers have investigated wind energy potential in Iran. These technical and economical investigations cover the different provinces of Iran where the climate changes significantly [12,13,14,15,16]. These studies investigate the wind energy in Iran while employing two-parameter Weibull distribution and demonstrated the suitability of the method for Iran.

Alhmoud and Al-Zoubi [17] suggested an IoT system to convert the data retrieved from the wind energy into actionable information to improve the performance of the wind turbine, driving the wind energy cost down and reducing the risk. The authors proved that the suggested system can be used to increase the reliability of measurement strategies, accuracy monitoring, and assurance of quality.

The demand for Artificial Intelligence (AI) and Internet of Things (IoT) have been also increased in recent years in the field of alternative energy resources [18,19,20]. For instance, the authors in [21, 22] suggested hybrid approaches of clustering-based probabilistic decision tree and deep learning to forecast wind power on large scales. Khan et. al. suggested a hybrid approach of clustering-based probabilistic decision tree to forecast the wind power efficiently. The authors used wind data to extract playing level fields for each feature after applying normalization and standard deviation. K-means clustering algorithm is applied to classify the samples into several groups that contain historical wind data. Various comprehensive comparisons are discussed showing the accuracy results of their research [21].

Khan et. al [22] proposed a new hybrid approach of principal component analysis (PCA) and deep learning-based research to uncover the hidden patterns using wind data and forecast the power of the wind. The authors applied PCA to wind data to extract the hidden wind data features and to achieve meaningful information. A TensorFlow framework is used to forecast wind power considering features and followed by the tuning of deep learning parameters to reach the optimal results to predict the wind signals and their approach has been applied to datasets from the National Renewable Energy Laboratory (NREL). The results of their research showed that they can accurately predict wind power using a span ranging from hours to years.

Batcha and Geetha [23] prepared a survey paper and they mentioned that AI can be used when the calculations are decoding gigantic knowledge arrangements to the Personal Computers, they can function without the need of specific programming and be utilized in Probabilistic Energy Forecasting. The authors examined the study of the IoT for a significant duty in solar-powered vitality, and the approach of AI is discussed and utilized in solar-oriented vitality.

Almonacid-Olleros et al. [24] suggested a new IoT architecture with machine learning paradigms in photovoltaic systems to forecast output energy. The authors evaluated several machine learning algorithms such as Linear regression, k-Nearest Neighbors, Support Vector Machines and Random Forest, to follow the behavior and energy production of a photovoltaic system in conjunction with the ambient data which retrieved by IoT environmental devices. The authors improved the results, considering the machine learning algorithms and compared their results with the state of art analytical model, and achieved significant differences in learning time as well as the performance.

Another research has been made by Yeh et al. [25] in long cycle maintenance prediction of wind turbines using machine learning models. The authors built a hybrid network to train the predictive model based on convolutional neural networks and support vector machines. The results showed that the prediction of their proposed method achieves high accuracy, and it is used to drive up the efficiency of wind turbine maintenance.

Akyuz and Demircan [26] proposed a research work based on IoT with Cloud Computing to monitor the wind turbine. The authors measured the parameters that are crucial for the performance evaluation of small power wind turbine and the measured data are used to evaluate the performance of the suggested system to avoid errors. The authors used the parameters such as wind speed, air temperature, battery voltage, current and recorded the data on a data logger. Microsoft Azure cloud computing has been used to record the data.

Ahmadi et. al [42] developed three models used to investigate the impact of input data on forecasting accuracy. The first model was trained with the average and standard deviation of wind speed values measured at a height of 40 m with a 10-min sampling time. To evaluate the impact of sampling time on model performance, a second model was trained with wind speed values measured at a height of 40 m with 1-h, 12-h, and 24-h sampling times.

3 Site description

Kerman province attracts researchers due to its historical and agricultural importance. Therefore, locations such as Kerman city [14] and Shahrbabak [27] have been investigated in terms of wind energy potential. This study investigates the feasibility of harvesting wind energy in Bam city. Kerman Province is the largest province of Iran located in the southeast of the country. The province covers 11% of the land area of Iran with an area of 183,285 km2 and a population of 3 million. The climate of the province varies from north to south. The North and central region of this province have a dry and moderate climate. On the other hand, the weather is warm and relatively humid in the southern part of the province. Most of the province is largely steppe or sandy desert, although there are some oases where dates, oranges, and pistachios are cultivated which the province is famous for. Bam is one of the biggest cities in Kerman province, houses “Arg-e Bam” the largest adobe building in the world, which is listed by UNESCO as part of the World Heritage Site. Despite the long history and the World's renowned Pistachios, the region still suffers from the devastating earthquake in 2003 which destroyed the city of Bam and the Citadel and its environs [28].

4 Wind data

The wind monitoring mast of Bam is located at 28º88ʹN 58º59ʹE and installed by The Renewable Energy Organization of Iran (SUNA) to measure the meteorological properties of the region. The mast height is 80 m and different sensors were mounted on three different heights of 40, 60, and 80 m. Data have been acquired by sensors such as wind anemometer, wind vane, temperature, radiation, and humidity sensors. The wind data were captured using a cup generator anemometer and acquired at 10 min time intervals from sensors mounted at three different heights. Several parameters have been acquired including wind speed, wind direction, temperature, humidity, and global radiation for the region. Data logger device reports maximum, minimum, average, and standard deviation values of wind speed together with average and standard deviation values of wind direction for each time interval of 10 min. Other properties such as air temperature, relative humidity, and global radiation are also available. In this study, the meteorological data acquired for one year from 1/01/2015 to 21/12/2015. This location has been chosen for investigation due to proximity to residential area and agricultural area. Thus, such a location provides the opportunity for electricity generation for both irrigation and residential use.

5 A Framework for wind energy harvesting in IoT (WHIoT)

The employed methodology for the wind energy harvesting in IoT (WHIoT) has been demonstrated in Figs. 1 and 2. Figure 1, represents the flowchart of the process.

Fig. 1
figure 1

Flowchart of the process of the present study

Full size image
Fig. 2
figure 2

Architecture diagram of main components and implementation of IoT

Full size image

The importance of the suggested framework is that the system can be followed by using the wind energy measuring equipment and the by the system administrators considering the input data from the acquisition devices. The equipment and sensors can be used for measuring the wind speed, where the system is followed by an AI model using the various input parameters and the specifications of machine specifications.

During data collection as the first stage, wind speed data have been acquired and also location terrain including latitude, longitude, and the elevation of the sites have been taken into consideration. The wind speed data have been formatted to correspond to a typical meteorological year. Then, the data have been modeled by employing a well-known method of Weibull distribution function. The next step is to select the number of commercially available wind turbines to assess their technical performance and economic feasibility to propose suitable wind turbines for wind energy development in the site. The suggested system is made up of three phases. The data collection phase includes data acquisition from the wind anemometers. In the wind data modeling, the wind speed data are considered and modeled by employing a two-parameter Weibull distribution function, as well as wind turbine selection based on the site wind speed data. All the data are stored in cloud storage to be used for the performance evaluation of the wind turbine where technical and economical evaluation is considered as two parts for implementation in industry. Figure 2 demonstrates architecture diagram of main components and implementation of IoT.

Wind energy potential can be estimated by employing statistical analysis. Therefore, statistical distribution parameters can be employed to compute wind energy characteristics such as mean speed, standard deviation, wind energy, and power density. The accuracy of the parameter estimation has a significant role in the selection of a suitable wind turbine for the investigated site to maximize wind farm electrical energy output and minimize the cost of energy generation. In this process, the number of acquired data, the time interval between data acquisition, the goodness of the fit test for the distribution of the parameters, and evaluation criteria of statistical data are key parameters affecting the performance of the distribution parameter estimation methods. Various probability density functions such as Weibull, Beta, Rayleigh, Gamma, Gaussian, and lognormal distribution can be employed to describe wind speed frequency. Weibull distribution is one of the common methods due to its simplicity, which can describe wind speed distribution with appropriate accuracy. The Weibull probability density function is defined as [29, 30]:

$$ f_{w} \left( v \right) = \frac{k}{c} \left( \frac{v}{c} \right)^{k - 1} exp\left( { - \left( \frac{v}{c} \right)^{k} } \right) ,$$
(1)

where \(f_{w} \left( v \right)\) is the wind speed probability for wind speed \(v\). Also, \(k\) and \(c\) are shape parameter (dimensionless) and scale parameter (m/s), respectively, and can be evaluated as follow [31, 32]:

$$ k = \left( {\frac{\sigma }{{\overline{v}}}} \right)^{ - 1.086} $$
(2)
$$ c = \frac{{\overline{v}}}{{{\Gamma }\left( {1 + 1/k} \right)}}, $$
(3)

where \(\overline{v}\) is mean of wind speed for a specified period and \(\sigma\) is the standard deviation of wind speed and can be evaluated as follow:

$$ \overline{v} = \frac{1}{n} \mathop \sum \limits_{i = 1}^{n} v_{i} $$
(4)
$$ \sigma = \left[ {\frac{1}{n - 1} \mathop \sum \limits_{i = 1}^{n} (v_{i} - \overline{v})^{2} } \right]^{0.5} $$
(5)

These parameters also can be evaluated as follows [33]:

$$ \overline{v} = \mathop \int \limits_{0}^{\infty } vf_{w} \left( v \right)dv = c {\Gamma }\left( {1 + \frac{1}{k}} \right) $$
(6)
$$ \sigma = \sqrt {c^{2} \left[ {\Gamma \left( {1 + 2/k} \right) - \left[ {\Gamma \left( {1 + 1 / k} \right)} \right]^{2} } \right]} ,$$
(7)

where \(\Gamma \left( x \right)\) is the gamma function and can be computed as follows [31]:

$$ \Gamma \left( x \right) = \mathop \int \limits_{0}^{\infty } exp\left( { - u} \right) u^{x - 1} dx $$
(8)

Wind power per unit area can be evaluated as follow:

$$ \frac{P}{A} = \frac{1}{2}\rho v^{3} \left( {W/m^{2} } \right) ,$$
(9)

where \(A\) is the perpendicular area, \(\rho\) is the density and \(v\) is the wind speed.

In addition, the Weibull probability density function can be employed to evaluate wind power density as follow [34]:

$$ \frac{P}{A} = \frac{1}{2}\rho \mathop \int \limits_{0}^{\infty } v^{3} f\left( v \right)dv = \frac{1}{2}\rho c^{3} {\Gamma }\left( {1 + \frac{3}{k}} \right) $$
(10)

Hence, the wind energy density over a specific period, T, and Capacity factor,\( C_{f}\), can be evaluated as [35, 36]:

$$ E = PT $$
(11)
$$ C_{f} = \frac{{E_{out} }}{{E_{r} }} $$
(12)

Capacity factor \( C_{f}\) is a very significant parameter of the productivity of wind turbines. The parameter represents the fraction of the total energy delivered over a period,\( E_{out}\), divided by the maximum energy that could have been delivered if the turbine was operating at maximum capacity over the entire period, \(E_{r} = \left( {365 days} \right) \times (24hr/day) \times P_{r}\).

To investigate the suitability of the fitted Weibull distribution, the coefficient of determination (R2) is computed. The coefficient also called the square of the Pearson correlation coefficient, shows the goodness of fit of probability density functions by evaluating the square of the empirical correlation between predicted wind speed and observations values. The coefficient ranges between the value of 0 to 1 to measure the estimation ratio of the distribution model. As the coefficient approaches 1, It demonstrates the best estimate [37, 38].

$$ R^{2} = \frac{{\mathop \sum \nolimits_{i = 1}^{N} \left( {y_{i} - z_{i} } \right)^{2} - \mathop \sum \nolimits_{i = 1}^{N} \left( {y_{i} - x_{i} } \right)^{2} }}{{\mathop \sum \nolimits_{i = 1}^{N} \left( {y_{i} - z_{i} } \right)^{2} }} ,$$
(13)

where \(z_{i}\) is the mean actual wind frequency percentages in time step i.

Economic feasibility is an important factor for potential investors. The economic feasibility measures the economic value of the project to determine the most feasible option. In this study, the feasibility of wind harvesting has been investigated by evaluating the unit cost of electricity. Operation and maintenance cost can express as a percentage of initial investment as [29, 39]:

$$ C_{OM} = mC_{I} $$
(14)

, where \(C_{OM}\) is operation and maintenance cost and \(C_{I}\) is the initial investment. Then the unit cost of energy can be expressed as:

$$ UCE = \frac{{C_{I} \left( {CRF + m} \right)}}{{E_{r} }} ,$$
(15)

where capital recovery factor (CRF) can be computed as follows:

$$ CRF = \frac{{d\left( {1 + d} \right)^{t} }}{{\left( {1 + d} \right)^{t} - 1}}, $$
(16)

where t is useful life cycle of the turbines and d is discount rate.

6 Results and discussions

In this section, we assess the proposed WHIoT approach via extensive scenarios. For wind energy harvest planning, it is essential to analyze wind speed data. Therefore, to demonstrate the wind energy potential average monthly and seasonal wind speed will be investigated and the wind data modeled by employing the Weibull distribution function. Further investigation was conducted on diurnal mean wind speed for energy demand analysis and wind rose diagram for site planning. The section concluded with remarks regarding the feasibility of harvesting wind energy from the investigated site by evaluating technical and economic feasibility metric indicators.

6.1 Monthly mean wind speed

Table 1 illustrates the monthly mean wind speed measured at three heights of 40, 60, and 80 m. The peak monthly wind speed average of 8.48 m/s was observed in July. On the other hand, the lowest monthly wind speed average occurred in December with a value of 3.92 m/s.

Table 1 Monthly mean wind speed for three heights of 40, 60, and 80 m
Full size table

Variation of seasonal wind speed is presented in Table 2 to provide additional information regarding wind profile in Bam city. It can be concluded that while summer exhibit the peak seasonal mean speed, winter exhibit the lowest seasonal mean wind speed.

Table 2 Seasonal mean wind speed variation measured at three heights
Full size table

Table 3 illustrates monthly mean wind speed and Weibull parameters for three different heights.

Table 3 Monthly mean wind speed and Weibull parameters for different heights
Full size table

As discussed previously, the two-parameter Weibull distribution function employed in this study. The shape parameter, k, demonstrates skewness rate of Weibull curve. Therefore, the curve is more skewed toward higher wind velocities for the higher values of k (2 < k < 3). On the other hand, values of k in range of 1 < k < 2 means the curve is more skewed toward lower wind velocities. Such observation indicates that lower wind velocities are more probable to happen. As it can be seen from the table, the variation of the shape parameters at a height of 80 m was in a range of 1.32 and 2.77. The scale parameter, c, illustrate the peak value of the Weibull curve. Therefore, higher values for c demonstrate a distribution over a wider range [4]. From the observed data, it can be seen that the scale parameters were observed to be in a range of between 4.91 and 9.54 m/s.

Table 4 illustrates Monthly power density and energy density for three heights in Bam. From the table, it can be seen that the peak values of power density and energy density occur during summer.

Table 4 Monthly power density and energy density for different heights
Full size table

The highest values of power density and energy density are observed in July with values of 547 \(\left( {W/m^{2} } \right){ }\) and 407 \(\left( {kWh/m^{2} } \right)\), respectively. Bam region has a warm summer with a mean temperature of 40° during July. As a result of such weather, the energy demand for cooling purposes is at a peak during this period. Thus, it can be concluded the wind energy can be coupled with another source of energy to meet the additional energy demand during the summer season. On the other hand, the minimum wind power was observed during November with values of 110.4 \(\left( {W/m^{2} } \right){ }\) and 79.5\(\left( {kWh/m^{2} } \right)\), respectively.

The diurnal mean wind speed of Bam is illustrated in Table 5. From the table, the hours of the day which wind speed is at the peak are apparent. The peak means wind speed occurred at an interval of 14:00–20:00 throughout the year.

Table 5 Diurnal mean wind speed for three heights of 40, 60, and 80 m
Full size table

6.2 Wind speed distribution

Annual variation of the mean value of Weibull wind speed frequencies for Bam is illustrated in Figs. 3, 4, and 5 for three heights. Wind speed distribution is depicted with the Weibull distributions in the figures. Weibull parameters were estimated by employing three methods, namely maximum likelihood, Least squares, and WAsP algorithm. It can be seen from the figure that the statistical distribution methods for modeling real situations are close to each other. The wind speed bin of 3 to 25 m/s is crucial in the modeling of actual data the energy production range of commercial wind turbines lies within this range. From the figure, it is apparent that the Weibull distributions show good agreement with actual distribution data.

Fig. 3
figure 3

Wind speed distribution at 80 m

Full size image
Fig. 4
figure 4

Wind speed distribution at 60 m

Full size image
Fig. 5
figure 5

Wind speed distribution at 40 m

Full size image

The performance and accuracy of these methods are investigated by performing an error analysis. Therefore, the coefficient of determination (\(R^{2}\)) is evaluated for all models and presented in Table 6. For a good model, the value of \(R^{2}\) approaches to unity. The results demonstrated the suitability of all three models for approximating the measured wind data of Bam city.

Table 6 Wind speed distribution error analysis at three heights
Full size table

Figure 6 illustrates the cumulative wind speed distribution for three heights. Cumulative distribution demonstrates the probabilities of a wind speed greater than a specific value. It can be observed that the probability of a wind speed at height 80 m is greater than other heights as the wind speed increases with height. It is apparent that the at heights of 40, 60, and 80 m wind speed greater than 4 m/s has a probability of 58%, 61%, and 64% of the year, respectively. This is a significant wind speed, as most commercial wind turbines have a cut-in speed of 4 m/s.

Fig. 6
figure 6

Cumulative density of wind speed at three heights for Bam

Full size image

6.3 Wind shear profile

The vertical wind shear profile of Bam city is illustrated in Fig. 7. From the figure, it can be observed that the wind speed suppresses the speed of 6 m/s at a height of 90 m. The power-law exponent (α) as well as a surface roughness (\(z_{0}\)) were computed with values of 0.179 and 0.217 m, respectively.

Fig. 7
figure 7

Vertical wind shear profile for Bam

Full size image

6.4 Wind rose diagram

Wind direction is an important parameter to obtain an optimum position for a wind farm.

The wind direction can be affected significantly by the geographical feature of the region and also atmosphere circulation. A wind rose diagram is a useful diagram that illustrates wind speed frequency and wind direction altogether in one diagram. Thus, the relative wind speed and its corresponding wind direction can be shown in a wind rose diagram. Figure 8 demonstrates the wind rose diagram for Bam, measured at a height of 78 m. the graph plotted by employing Windographer software for 24 sectors with an arc of 15°. From the figure, it is apparent that the prevailing wind direction from North West.

Fig. 8
figure 8

Wind rose diagram of Bam city

Full size image

6.5 Economic evaluation

To evaluate the wind energy potential and investigate the feasibility of harvesting the wind power in Bam, six different wind turbines are selected and their performance is investigated in terms of Annual Energy Production (AEP), Capacity Factor (CF), and Unit Cost of Energy (UCE).

Windographer software has been employed to compute capacity factor \(C_{f}\) and annual energy output AEP of the selected wind turbines. Several parameters such as awake, power curve hysteresis, and connection grid downtime were considered to evaluate the production losses due to wind turbine downtime with 10% production losses. To perform the economical assessment, initial costs,\({ }C_{I}\), including costs such as transportation, installation, and grid integration are taken to be 40% of the turbine cost [34, 40]. Also, annual operating and maintenance costs, \(C_{OM} ,\) together with land are assumed to be 5% of the wind turbine cost [34]. Other parameters such as discount rate and expected useful life cycle of the turbines are taken as 10.5% [41] and 20 years, respectively.

Table 7 shows specifications of the selected turbines, AEP, \(C_{f}\) and UCE. In terms of AEP Endurance E-3120 (Grid-connected) and Proven 2.5 kW (Off-grid) produces the highest amount of energy. On the other hand, \(C_{f}\) is an important parameter for analyzing wind turbine performance which also affects the cost of electricity generation. From the presented results, it can be seen that wind turbine Proven 15 kW demonstrates the highest efficiency in terms of a capacity factor with a value of 28.67%. From the table, it is apparent that the unit cost of energy in Bam ranges from 0.491 to 1.512 (USD/kWh). The grid-connected Proven 15 kW demonstrates the cheapest energy production.

Table 7 Technical specification and performance of the selected wind turbines
Full size table

7 Conclusion

Electricity generation from wind energy as clean and sustainable energy increasing significantly. Hence, to harvest energy by employing wind turbines, it is required to evaluate the wind characteristics of the region. In this study, the wind potential of Bam city has been investigated by employing wind speed data acquired by the Renewable Energies Organization of Iran for one year with ten minutes intervals. The annual wind speed at 80 m was observed to be 5.92 m/s with maximum and minimum monthly wind speed average that occurred in July and December, respectively. Hence, in July power density and energy density reached a peak value of 547 \(\left( {W/m^{2} } \right){ }\) and 407 \(\left( {kWh/m^{2} } \right)\), respectively. On the other hand, the minimum seasonal wind speed occurred was observed during winter. Such variation provided an opportunity for harvesting wind energy to meet the domestic demand of the region. Thus, during warm summer with peak demand of energy for cooling purposes, wind energy can be utilized to meet the additional energy demand.

The result demonstrated the prevailing wind direction was observed from North West in this region, which must be considered for development future wind farm. Finally, the performance of 6 Grid-connected and Off-grid small-scale wind turbines has been assessed. The feasibility investigation revealed that the UCE in Bam for small-scale wind turbines ranges from 0.491 to 1.512 (USD/kWh). Future work will include the IoT System to communicate and collect data with the server to be used for further processing of data. In this way, it is envisaged to investigate the possibility of improving wind energy production by integrating the classical production site with new IoT systems, and the application of the expert system, as well as statistics to reach better results and performance.