You are currently viewing a new version of our website. To view the old version click .
MDPIMDPI
Search all articles
Energies
Download PDF
  • Feature Paper
  • Article
  • Open Access

9 December 2025

Research and Analysis of the Impact of Local Climatic Conditions on Wind Turbine Generation—Case Study

,
and
1
Department of Photonics, Electronics and Lighting Technology, Bialystok University of Technology, Wiejska 45A, 15-351 Bialystok, Poland
2
Department of Electrotechnics, Power Electronics and Electrical Power Engineering, Bialystok University of Technology, Wiejska 45A, 15-351 Bialystok, Poland
*
Author to whom correspondence should be addressed.
Energies2025, 18(24), 6429;https://doi.org/10.3390/en18246429
This article belongs to the Section K: State-of-the-Art Energy Related Technologies
Version Notes
Order Reprints

Abstract

The increasing number of wind turbines in sparsely populated areas poses significant challenges to the security of the power system. The need for centralized control of unstable sources such as wind turbines and PV installations makes it significantly more difficult to maintain operational stability. The subject matter of wind power plant research can be divided into three groups: wind prediction and productivity forecasts, optimization of the energy generation process, and the impact of power plants on the system and the environment. The subject matter of this article falls within the scope of research on the first group, namely, the impact of wind on actual power production. This study presents the results of a year-long investigation of the Enercon E48 wind turbine located in northeastern Poland. The influence of wind speed and direction on the actual turbine output was analyzed and compared with the manufacturer’s power curve. The findings indicate that the actual performance of the turbine exhibits greater variability than suggested by catalog data, with local conditions and seasonal effects exerting a significant influence on its efficiency.

1. Introduction

The share of renewable energy in total electricity production has been growing dynamically since 2000, reaching 45.3% in the European Union and 29.6% in Poland in 2024 (Eurostat data) (Figure 1). Among renewable energy sources worldwide, hydropower plants still generate the largest amount of electricity. However, the highest growth rates are observed in solar and wind power plants (Figure 2). Within the European Union and Poland, wind power plants remain the most significant renewable energy source (Figure 3 and Figure 4).
Figure 1. Share of electricity generated by renewables, Percentage of electricity produced from renewable sources, which include solar, wind, hydropower, bioenergy, geothermal, wave, and tidal. Data source: Ember (2025); Energy Institute-Statistical Review of World Energy (2025) [1].
Figure 2. Renewable energy generation, World. ‘Other renewables’ refers to renewable sources including geothermal, biomass, waste, wave and tidal. Traditional biomass is not included. Data source: Energy Institute-Statistical Review of World Energy (2025) [1].
Figure 3. Renewable energy generation, European Union. ‘Other renewables’ refers to renewable sources including geothermal, biomass, waste, wave and tidal. Traditional biomass is not included. Data source: Energy Institute-Statistical Review of World Energy (2025) [1].
Figure 4. Renewable energy generation, Poland. ‘Other renewables’ refers to renewable sources including geothermal, biomass, waste, wave and tidal. Traditional biomass is not included. Data source: Energy Institute-Statistical Review of World Energy (2025) [1].
The share of wind-generated electricity remains higher than that of solar, despite the rapidly increasing installed capacity of PV systems, which in many cases already exceeds that of wind power plants. The total installed capacity in the European Union at the end of 2024 amounted to: solar—334 GW, wind—269.7 GW. In Poland, the corresponding figures were 20.2 GW for solar and 10.8 GW for wind.
The amount of electricity generated from individual renewable sources depends strongly on local geographic and meteorological conditions (Figure 5) [2,3]. In 2024, the share of wind energy in total electricity generation was 17.5% in the European Union, 14.5% in Poland, and as much as 57.9% in Denmark. A comparison of several selected countries and regions is presented in Figure 5.
Figure 5. Share of electricity production from wind. Measured as a percentage of total electricity produced in the country or region. Data source: Energy Institute-Statistical Review of World Energy (2025) [1].
The extensive integration of variable renewables increasingly affects the stability of power systems. This can result in a variety of challenges, such as deterioration in power quality parameters or difficulties in balancing electricity supply and demand. Due to insufficient storage capacity and transmission capabilities of power grids, curtailment measures are increasingly being imposed on renewable energy plants. For example, in the first half of 2025 in Scotland, curtailments affected as much as 37% of forecasted wind farm production.
The growing number of wind turbines in sparsely urbanized areas creates additional challenges for power system security [4]. The necessity of centralized control of variable sources, such as wind turbines and PV systems, significantly complicates the maintenance of operational stability [5].
Issues related to modeling and assessing wind parameters are a key area in the development of power forecasting methods [6,7,8,9]. The literature on optimizing the energy generation process [10,11,12,13] emphasizes the importance of accurately mapping the aerodynamic conditions of turbines. In turn, studies on the functioning of wind farms in the power system [14,15,16] draw attention to the need to integrate accurate wind forecasts with grid stability models. In the context of environmental analyses, researchers emphasize the impact of wind variability and turbine operation on local ecosystems [4,17].
As renewable energy sources become increasingly important, accurate forecasting of their production is essential [18]. Forecasting wind energy production is particularly difficult because it depends not only on wind speed but also on a number of additional factors. Turbulent phenomena can significantly modify the instantaneous power generated by a turbine, which makes it very difficult to create stable predictive models [19,20]. At the same time, studies on the impact of air density [21,22] indicate its key role in determining the efficiency of converting kinetic energy into electricity. Analyses devoted to wind direction [23,24,25] draw attention to the need to take into account its variability in the short and medium term, as it affects both the load on the turbine and its efficiency. The variation in wind speed at different heights also significantly affects the operation of turbines and, consequently, the accuracy of forecasts [26,27]. Traditionally, power generation forecasts for wind turbines are based on the manufacturer’s power curve, which relates mean wind speed to generated power [28]. However, studies have shown that such a simplified approach can lead to forecasting errors of 5–10% [29]. These factors can significantly affect the actual turbine performance and cause deviations from expected results [30]. Consequently, more advanced models are being developed to transform atmospheric data into realistic forecasts of wind energy production [31]. This enables wind energy to be more effectively utilized as a stable component of the energy system.
The power curve is certified in accordance with the guidelines of the International Electrotechnical Commission (IEC) by the turbine manufacturer [32]. It can be used as the main indicator when comparing wind turbine types during the design of wind farms [33]. Careful assessment and selection of wind turbine characteristics adapted to the wind regime at a given location allows for the optimization of energy production from a wind farm [34,35]. First and foremost, the size of the wind turbine should be adapted to its location, while taking into account its availability, reliability and warranty period [36,37].
The power curve as a function of wind speed plays a significant role in several key applications within the wind energy sector. These applications include, among others, wind turbine selection, estimation of the capacity factor, wind energy assessment and forecasting, as well as condition monitoring. Although this subject has already been described in numerous publications, with power curves derived both from measurements and from various types of models, the authors have not encountered studies addressing the influence of wind direction, time of day, and season on electricity production. This article aims to evaluate the impact of wind speed and direction on the actual power generation of the Enercon E48 wind turbine. In addition, the influence of time of day and season on the generated power was analyzed.
The structure of the article is as follows: Chapter 2 presents the technical parameters of the wind turbine under investigation and the relationships used in the analysis of the results. Chapter 3 presents and discusses the results of the research. Finally, the main conclusions are summarized in Chapter 4.

2. Materials and Methods

The power curves provided in turbine catalogs are based on a theoretical model consistent with IEC 61400-12-1 [38], which can be expressed as
P = 1 2 ρ A ν 3 C p ,
gdzie: ρ—air density; A—rotor swept area; ν—wind speed; Cp—power coefficient (not exceeding 0.593 [39]).
To determine the actual power curve of a turbine as a function of wind speed, simultaneous measurements of wind speed and turbine output are typically taken at 10 min intervals [38]. Power output is recorded using the SCADA system, while wind speed can be obtained from a meteorological tower, measured with nacelle-mounted anemometers [40]. However, wind speed measurements taken at the nacelle are subject to errors caused by rotor-induced flow disturbances [41]. Data on wind speed and generated power often contain numerous anomalies resulting from sensor errors, communication disruptions, and operational situations such as production restrictions or maintenance periods. Such irregularities reduce the quality of analyses and significantly impair the effectiveness of wind power forecasting [42]. The application of a Lidar system for wind speed measurement enables improved turbine control and minimization of losses, which can translate into increased energy production from the wind turbine [43]. Wind turbine power curves generally consist of three main regions: the cut-in region, the working region, and the rated power region [44].
The subject of the study was the recording and evaluation of parameters describing the power generated by the Enercon E48 (Figure 6) wind turbine as a function of wind conditions. The turbine was connected to a 20 kV power grid via a 900 kVA oil transformer (21/0.4 kV; Dyn5; ΔUz% = 6%), located in northeastern Poland (Podlaskie Voivodeship, Suwałki County). The turbine was located on a small hill, at a distance of no less than 1900 m from residential buildings and more than 1450 m from the forest (with a few isolated trees and shrubs located closer, none exceeding 5 m in height). The measurements were carried out continuously over a period of one year (2024). Wind parameters and turbine output power were recorded around the clock at 10 min intervals. The wind measurement device was installed directly on the nacelle of the investigated wind turbine and was synchronized with the instrument measuring the power output of the plant. In order to reduce the error resulting from disturbances introduced into wind speed measurements due to turbulence and aerodynamic shadowing behind the rotor, the anemometer on the nacelle was calibrated against reference measurements using the nacelle transfer function (NTF—Nacelle Transfer Function) in accordance with IEC 61400-12-2 [38]. The analyzed turbine used a SCADA system, which collected data on generator operating parameters, rotor speed, atmospheric conditions, and wind direction and strength, and then processed it to obtain the necessary data. The actual power delivered to the grid was measured using electrical systems that recorded voltage, current, and power factor. The basic technical data of the investigated wind turbine are presented in Table 1.
Figure 6. The Enercon E48 wind turbine under investigation.
Table 1. Basic technical parameters of the investigated wind turbine.
The study also included a basic statistical analysis of the measured values, described by the following parameters (notations used: s—standard deviation; xᵢ—individual i-th value; x ¯ —arithmetic mean; n—sample size; SE—standard error of the mean; CI—confidence interval):
  • Standard deviation, which indicates how much the individual values deviate from the mean:
s = 1 n − 1 ∑ i = 1 n x i − x ¯ 2
  • Standard error of the mean, which determines the uncertainty of the mean estimate, i.e., how much the sample mean may differ from the population mean:
S E = s n
  • Confidence interval describes type A measurement uncertainty in statistical data, showing the range of values within which the true value of the measured quantity lies with a specified probability. A confidence interval of 95% was adopted for the analyzed data:
C I 95 % = x ¯ ± 1.96 S E = x ¯ ± 1.96 s n
  • Kurtosis describes how strongly the data are concentrated around the mean and how heavy the tails of the distribution are (leptokurtic, platykurtic, or normal distribution):
K r = n n + 1 n − 1 n − 2 n − 3 ∑ x i − x ¯ s 4 − 3 n − 1 2 n − 2 n − 3 − 3
  • Sample variance describes how much the values differ from the mean (calculated as the square of the standard deviation).
  • Skewness indicates the asymmetry of the data distribution relative to the mean (whether most values lie above or below the mean).
  • Mode, i.e., the most frequently occurring value in the dataset.
The statistical parameters introduced are not only used to describe the data, but also enable the identification of asymmetries, variability, and extreme events, which are key to the proper interpretation of the impact of local environmental conditions on turbine operation. In addition, confidence intervals allow the stability of the power plant’s output to be assessed as a function of wind speed and determine whether the difference from the manufacturer’s curve is significant.

3. Results and Discussion

The characteristic power curves of the wind turbine obtained from measurements and those provided by the manufacturer are shown in Figure 7. The orange line represents the manufacturer’s curve, i.e., the expected active power output as a function of wind speed. The measurement data, marked as blue points, reflect the actual power generated by the turbine. In the lower wind speed range, below 4 m/s, both the theoretical model and the measurements indicate very low or zero power output. This is consistent with the principle that turbines require a minimum airflow to start operation. However, several points with positive power can be observed in this range, which may result from the inertia of the turbine—while the wind speed had already decreased, the rotor continued spinning from the momentum of a previous gust. In the range of approximately 4–10 m/s, the turbine begins to produce increasing amounts of power. The manufacturer’s curve rises steeply, whereas the measured data points show noticeable dispersion. Interestingly, most of these points are above the manufacturer’s curve, suggesting that in practice the turbine generates more power than expected. The considerable scatter of measured values demonstrates that power generation depends not only on wind speed but also on other factors, such as air density, flow stability, control system efficiency, or the technical condition of the device. In the higher wind speed range, i.e., above 13 m/s, the manufacturer’s curve reaches its maximum value—the turbine should operate at full power, reaching 810 kW. The measured points are clustered around this value, but they still tend to exceed the manufacturer’s specification. It is also noteworthy that the turbine achieves its rated power at a wind speed nearly 2 m/s lower than indicated by the manufacturer. Furthermore, the maximum power achieved by the turbine exceeds the catalog value (810 kW) by nearly 25 kW. This contrasts with the findings of Zou et al. [38], who reported that manufacturer power curves often overestimate real output, with deviations of up to 10% at wind speeds around 13 m/s.
Figure 7. Power output of the investigated wind turbine as a function of wind speed (measured data vs. manufacturer’s curve).
A wind turbine can exceed its rated power in several well-documented situations resulting from both atmospheric conditions and the dynamics of the system itself. Firstly, the power available in the wind depends on air density, which increases at low temperatures, high pressure, and low humidity. Under such conditions, especially in winter, the turbine can receive significantly more energy at the same wind speed, leading to temporary increases in output power above the rated value. Secondly, gusts and turbulence in the wind cause sudden jumps in the speed of the wind stream. The blade pitch control system does not respond immediately, so before the system has time to limit the power, the generator may produce more than 800 kW for a short time. The kinetic energy stored in the rotor plays an additional role. During sudden changes in operating conditions, such as when switching from power reduction to normal operation or during a sudden increase in wind speed, the rotor can transfer more energy to the generator than would be expected under steady-state conditions. All these factors mean that, in practice, an 800 kW turbine can temporarily achieve a power output of 820–850 kW, especially in non-stationary, dynamic wind conditions and at low ambient temperatures.
Comparison of the theoretical and measured curves shows that, despite the general consistency of trends, the actual operating conditions of the turbine are more complex and variable [45]. Such analysis provides insights into the turbine’s efficiency, highlights potential sources of losses, and can support optimization of power system operation. For the investigated turbine, generation should be improved (possibly through optimization of blade pitch control) at low wind speeds (<6.5 m/s), where most measured values fall below the manufacturer’s assumptions. It should also be emphasized that the recorded actual power curve as a function of wind speed does not exhibit the significant anomalies reported by Bilendo et al. [46] in analogous measurements. Anomalies in power curve measurements are mainly caused by rotor blade surface degradation, turbine misalignment with the wind direction, and incorrect blade pitch angle settings relative to wind speed [47]. In the analyzed turbine, a deviation of the actual generated power can be observed both at low wind speeds (3–6 m/s) and at wind speeds at which the turbine should be approaching its rated power (9–13 m/s). At higher wind speeds, the deviation from the manufacturer’s curve is positive (the turbine generates more power than specified by the manufacturer). At lower wind speeds, however, the deviation is negative—the actual power generated by the turbine is lower than expected. This indicates that at low wind speeds, the blades are not optimally pitched, which does not ensure sufficient lift to drive the generator. These settings should be verified by the turbine maintenance service.
Figure 8 shows the results of the statistical analysis of the relationship between the power generated by the investigated turbine and wind speed.
Figure 8. Graphical representation of the statistical analysis results showing the relationship between the power generated by the investigated turbine and wind speed.
The analysis of Figure 8 indicates a significant relationship between the power generated by the investigated wind turbine and wind speed, with particular attention to the dispersion of values observed across different ranges. At low wind speeds, the dispersion is small, meaning that the turbine predictably generates little or no power, remaining close to minimal values. As wind speed increases, the variability of results becomes more pronounced, reflecting the changing atmospheric conditions and the irregularity of instantaneous airflow. The largest dispersion is observed in the medium wind speed range, where power output increases dynamically, but substantial differences between individual measurements are evident. This highlights the high sensitivity of the turbine to short-term variations in wind intensity. At very high wind speeds, the dispersion decreases, which results from the action of regulatory mechanisms that limit power output to protect the turbine. Overall, the analysis demonstrates that while wind speed determines the general shape of the power curve, it is the dispersion of values within individual ranges that best reflects the dynamics and instability of turbine operation under changing environmental conditions.
The histogram presented in Figure 9 illustrates the frequency distribution of wind speeds at the turbine site. The results show that the most frequent wind speeds fall within the low-to-medium range.
Figure 9. Histogram of wind speed distribution at the investigated wind turbine site.
The histogram has an asymmetric shape, with a pronounced peak at moderate wind speeds (4.5–6 m/s) and decreasing frequencies for both higher and lower values. This indicates that winds at the site most often blow at moderate speeds, while very strong winds (>14 m/s) are rare (wind speeds above 19.7 m/s were never recorded). The distribution is particularly relevant for the investigated turbine, since at wind speeds below 6.5 m/s (which account for the majority of measurements), the turbine generates power below the manufacturer’s predicted values. Table 2 presents the statistical parameters describing the recorded wind speeds.
Table 2. Selected statistical parameters of the recorded wind speeds.
The statistical parameters presented in Table 2 provide a better understanding of the characteristics of the local wind regime. The positive skewness of the distribution (0.497) indicates the presence of a long right tail, which means that higher wind speeds occur less frequently but have a significant impact on the turbine’s instantaneous high power output. The kurtosis value (~1.02) confirms a flatter than normal distribution, consistent with the wide range of moderate speeds visible in the histogram (Figure 8). In turn, the modal speed value (5 m/s) and the median (5.8 m/s) prove that typical turbine operating conditions fall within the speed range below 6.5 m/s, i.e., in the zone where the power generated is lower than the catalog value. Combined with these statistical parameters, a more complete explanation is obtained as to why the turbine operated at a power significantly lower than the nominal power for most of the time.
In summary, the wind speeds show slight right-skewness, moderate variability, and a mild tendency toward outliers at higher values. The results differ from the statement in [48] that the most frequent winds should fall within 0.9–1.1 of the mean; here, the range extends from 0.5 to 1.5.
The frequency distribution of power output is shown in Figure 10. The histogram indicates a clear dominance of low power values—particularly in the 0–20 kW range, where the number of occurrences exceeds 9500. This suggests that the turbine often operates under suboptimal conditions, producing power far below its nominal capacity. This is surprising, given that the most frequent wind speeds (Figure 9) fall between 2.5 and 9.5 m/s, where output should be much higher than 20 kW.
Figure 10. Histogram of power output distribution of the investigated wind turbine.
As power increases, the number of cases systematically decreases, indicating that higher generation levels occur only sporadically. Power values above 550 kW are relatively rare, suggesting limited exploitation of the turbine’s full potential, as conditions favoring maximum output occur only briefly. The histogram is asymmetric, with a strong concentration near the lower end and a long tail extending toward higher values.
Statistical analysis of the generated power (Table 3) reveals a very strong asymmetry in the distribution. This is confirmed by the positive skewness (1.675) and high kurtosis (2.58), characteristic of distributions with a long right tail, typical for turbines operating in highly variable wind conditions. This means that most values are concentrated in the low power range, while rare but high values—associated with short-term favorable aerodynamic conditions—significantly increase the average power (162 kW). In addition, the discrepancy between the mean and the median (99 kW) and the fact that the modal value is 0 kW indicate that the turbine most often operated near the cut-in threshold. A comparison of these statistical parameters confirms that the variability of atmospheric conditions and the non-stationarity of the flow have a key impact on the efficiency of the turbine, which is not apparent from the average values alone. These extremes have a significant impact on the statistical parameters, in particular on the high standard deviation (~179 kW) and variance (~32,024). The maximum recorded power of 834 kW exceeded the rated power by more than 4%. Overall, the analysis indicates that power generation occurs mainly at low levels, but the system is technically capable of reaching full power under favorable conditions, which is consistent with previous studies [49].
Table 3. Selected statistical parameters of the recorded power output.
Figure 11 shows the daily variability of wind speed recorded by the measurement system installed at the wind farm. For this purpose, wind speeds occurring at that time were given six times for each hour. The values presented refer to the entire measurement year, which means that 365 wind speed values were given for each point.
Figure 11. Diurnal variability of wind speed recorded at the investigated wind turbine site.
The diurnal profile (Figure 11) shows that wind speed undergoes dynamic fluctuations rather than remaining constant. Most values fall between 1 and 18 m/s, with clustering in the 3–9 m/s range, suggesting a typical wind profile for the site. No clear day–night pattern is observed. Interestingly, studies in [50] indicated that wind speeds in Europe typically peak around noon.
Figure 12 presents the diurnal variability of the active power generated by the investigated wind turbine. The values are presented in the same way as wind speed (for every 10 min, the values generated by the power plant during the year are presented). Significant fluctuations in power output are observed throughout the day, resulting from variations in wind speed and the turbine’s response to instantaneous atmospheric conditions. The distribution of generated power is uneven, with periods of more intensive production alternating with phases close to the minimum operating threshold. No distinct daily pattern indicating predictable peak production hours can be identified. The observed variability is crucial for assessing the operational efficiency of the turbine, particularly regarding its responsiveness to rapid external changes. Such fluctuations imply the necessity of implementing intelligent control systems that enable dynamic adaptation of turbine operation, especially under low wind speed conditions. The power production observed in the investigated wind turbine markedly differs from the diurnal cycle reported for the Taïba Ndiaye wind farm [51], where a distinct midday decline (11 a.m.–5 p.m.) was recorded.
Figure 12. Diurnal variability of power output of the investigated wind turbine.
The analysis of Figure 13, which presents the relationship between the power generated by the investigated wind turbine and the hour of the day, indicates both variability in the dispersion of values and significant differences in mean and median levels. During the nighttime hours, particularly between 22:00 and 05:00, turbine power remains relatively high, with considerable dispersion. The medians in this period are close to each other, confirming stable wind conditions. In the morning hours, from 06:00 to 10:00, both the median and the dispersion gradually decrease, reflecting reduced atmospheric activity and less favorable conditions for power generation. The lower dispersion also indicates more stable airflow conditions. The highest power values are observed in the afternoon, between 11:00 and 16:00, suggesting that the turbine achieves the most favorable operating conditions during this interval. However, dispersion is also greatest in this period, and the presence of numerous outliers indicates sudden and difficult-to-predict changes in wind intensity. Peak energy production at these hours is disadvantageous both for the power grid and for the wind farm owner. This is because maximum photovoltaic generation occurs at the same time, resulting in increased grid voltage, network overloading, and consequently lower energy prices on the hourly electricity market.
Figure 13. Graphical representation of the statistical analysis results showing the relationship between the power generated by the investigated turbine and the hour of the day.
Figure 14 presents the scatter plot illustrating the relationship between wind speed and wind direction recorded at the investigated turbine site. For each wind direction, the wind speeds recorded during the analyzed year are presented. The distribution of data shows considerable variability in wind speed depending on direction, with values not evenly dispersed. A concentration of higher wind speeds (>14 m/s) can be observed in the western sector (~270°), suggesting that dominant, high-intensity winds occur from this direction. Measurement areas between northwest (315°) and east (90°) are characterized by lower speeds and lower point density, indicating weaker and less frequent winds from these directions. The highest point density (i.e., most frequent winds) is recorded for directions ranging from south (180°), through southwest (225°) and west (270°), to northwest (315°).
Figure 14. Scatter plot of wind speed variability as a function of wind direction at the investigated site.
The irregular and asymmetric distribution of points may be attributed to local topographic conditions, such as terrain relief, aerodynamic obstacles (e.g., buildings, tree lines), or the presence of natural wind corridors. The data in Figure 14 can be applied to optimize turbine placement and rotor orientation for maximum energy production. They help identify prevailing wind directions, whose higher frequency and greater intensity directly influence turbine efficiency. This is particularly important for site-specific analysis, since dominant wind directions differ across regions—for instance, in the Mediterranean basin, prevailing winds are predominantly northerly [52].
To verify the correctness of the generator’s response to wind speed variation, Figure 15 presents a scatter plot of wind direction (in degrees) against the active power output of the turbine.
Figure 15. Scatter plot of active power output as a function of wind direction at the investigated site.
The distribution of points in Figure 15 is non-uniform, with distinct clusters of higher power values observed in certain directions, suggesting dominant wind sectors. Similarly to wind speed, the power generated by the turbine during the analyzed year is presented for each degree of wind direction. In particular, the highest power outputs (exceeding 700–800 kW) occur in association with the western sector (with slight deviations to the north and south). This indicates that winds from these directions are the strongest and most stable, favoring maximum turbine performance. Conversely, northern sectors show significantly lower outputs and sparser point density, reflecting their limited contribution to energy generation. These observations are consistent with the wind direction variability shown in Figure 14.
Figure 16 shows the graphical interpretation of the statistical analysis results of the relationship between the power generated by the investigated turbine and wind direction.
Figure 16. Graphical representation of the statistical analysis results showing the relationship between the power generated by the investigated turbine and wind direction.
The analysis of Figure 16 illustrates the relationship between the power generated by the investigated wind turbine and wind direction, clearly indicating that airflow direction plays a crucial role in the operational efficiency of the device. The highest power values are obtained for winds blowing from the western and south-western sectors, which can be associated with the dominant circulation patterns in the region that favor stronger air movements. North-western winds also prove beneficial, although they exhibit somewhat greater variability. Eastern and south-eastern directions are less effective, resulting in lower power output, while southern winds perform even worse. The lowest output—more than twice as low as that of the most productive directions—is recorded for northern winds, most likely due to the infrequent occurrence of conditions favorable to such flows. The box-and-whisker plot further indicates the presence of outliers, which can be interpreted as the effect of occasional atmospheric phenomena, such as strong gusts from less typical directions. Overall, the analysis confirms that optimal turbine performance depends on prevailing winds from the western sector, whereas northern winds should be regarded as less favorable for energy production. This is particularly important from the perspective of wind turbine siting, as no obstacles should obstruct airflow in the prevailing wind directions.
Figure 17 presents the scatter distribution of monthly wind speeds. The dispersion of values allows identification of months with higher wind speeds and those with weaker winds. Notably, November, December, January, and April exhibit frequent wind speeds exceeding 13 m/s (the rated wind speed of the turbine). By contrast, summer and early autumn months (June–October) are characterized by lower wind speeds, typically below 10.5 m/s.
Figure 17. Distribution of average monthly wind speeds at the investigated site.
The observed seasonality is typical of a temperate climate and may be attributed to the influence of synoptic-scale pressure systems [53]. The clear increase in winter and decrease in summer have practical significance for turbine operation planning, highlighting periods of higher generation potential as well as windows of reduced production that can be allocated for maintenance.
The bar chart in Figure 18 shows the distribution of average monthly active power output of the turbine.
Figure 18. Distribution of average monthly active power output of the investigated wind turbine.
A pronounced seasonal pattern is evident. The highest average output (~275 kW) was recorded in January, significantly exceeding values in November and December despite comparable wind distributions. In February, March, and April, power generation follows the wind speed patterns shown in Figure 14. The lowest mean output (~75 kW) occurred in June. In subsequent months (August–October), output increases despite similar average wind speeds, suggesting greater stability of autumn winds compared with summer. This seasonal pattern confirms the winter peak in wind energy production and emphasizes its importance for strategic planning [53]. Such analysis is particularly valuable for forecasting peak production periods, optimizing maintenance schedules, managing grid integration and storage, and assessing the economic feasibility of wind projects under local climatic conditions.
Figure 19 shows the results of the statistical analysis of the relationship between the power generated by the investigated turbine and the month of the year.
Figure 19. Graphical representation of the statistical analysis results showing the relationship between the power generated by the investigated turbine and the month of the year.
The analysis of Figure 19 indicates a significant relationship between the power generated by the investigated wind turbine and the month of the year, clearly confirming the seasonal nature of wind energy production. A distinct variability of power output across different periods can be observed, with winter and early spring months characterized by higher values, which may be related to increased atmospheric activity and higher wind speeds. In contrast, during the summer period, the power generated by the turbine decreases markedly, suggesting less favorable anemometric conditions. The box-and-whisker plot also highlights the variability of power output within individual months, reflecting fluctuations in weather conditions on a shorter time scale. These results have practical significance, as they confirm the necessity of accounting for seasonality when planning the operation of wind turbines and may also provide a basis for further research on optimizing turbine performance under variable environmental conditions.

4. Conclusions

This study presented the results of a year-long investigation of the Enercon E48 wind turbine located in northeastern Poland, aimed at comparing the actual power output with the manufacturer’s curve and analyzing the influence of wind speed and direction on turbine performance. The average wind speed during the study period was approximately 5.94 m/s, with a median of 5.8 m/s, and the distribution was slightly asymmetric (skewness ~0.5). The mean active power output of the turbine was only about 162 kW, corresponding to ~20% of the rated capacity of 800 kW. The most frequently recorded value was 0 kW, while the median output was 99 kW, indicating that the turbine operated at minimal power for a significant portion of the time. Nevertheless, under favorable conditions, the turbine achieved and even exceeded its rated output: the maximum recorded power was 834 kW, i.e., 24 kW higher than the catalog value, with rated power reached at wind speeds about 2 m/s lower than specified by the manufacturer. The highest wind speeds (>14 m/s) and greatest power outputs (>700–800 kW) were observed mainly in the winter season, especially in January, when the average power generation reached ~275 kW, while the lowest mean value (~75 kW) occurred in June. The analysis confirmed that the most productive winds originated from the western sector, whereas winds from other directions were weaker and contributed less to energy generation.
Statistical analysis of wind speed and power generation distributions played an important role in interpreting turbine performance. Skewness and kurtosis values revealed the asymmetric and heavily tailed nature of both distributions, which helps to explain the predominance of low-power periods and the occasional attainment of values close to and exceeding the rated power. These results show that statistical parameters—properly linked to the physical mechanisms of turbine operation—are a valuable tool for assessing actual deviations from the manufacturer’s curve.
In summary, the actual operation of the turbine proved to be more variable than suggested by catalog data. Seasonality, prevailing wind direction, and local site-specific conditions had a significant impact on the efficiency of the power plant. The findings highlight the necessity for further optimization of turbine operating parameters, particularly at low wind speeds, in order to enhance efficiency under conditions that occur most frequently at this location.

Author Contributions

Conceptualization, J.K., Z.S. and G.H.; methodology, J.K., Z.S. and G.H.; software, Z.S. and G.H.; validation, Z.S. and G.H.; formal analysis, Z.S. and G.H.; investigation, Z.S. and G.H.; resources, Z.S. and G.H.; data curation, Z.S. and G.H.; writing—original draft preparation, J.K., Z.S. and G.H.; writing—review and editing, J.K., Z.S. and G.H.; visualization, J.K., Z.S. and G.H.; supervision, J.K.; project administration, J.K.; funding acquisition, J.K. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the Bialystok University of Technology as part of the teamwork WZ/WE-IA/3/2023.

Data Availability Statement

The data presented in this study are available upon request from the corresponding author.

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. Available online: https://ourworldindata.org/ (accessed on 18 August 2025).
  2. Kusznier, J. Influence of Environmental Factors on the Intelligent Management of Photovoltaic and Wind Sections in a Hybrid Power Plant. Energies 2023, 16, 1716. [Google Scholar] [CrossRef]
  3. Kusznier, J.; Wojtkowski, W. Impact of climatic conditions on PV panels operation in a photovoltaic power plant. In Proceedings of the 2019 15th Selected Issues of Electrical Engineering and Electronics (WZEE), Zakopane, Poland, 8–10 December 2019. [Google Scholar] [CrossRef]
  4. Magdy, G.; Shabib, G.; Elbaset, A.A.; Mitani, Y. A Novel Coordination Scheme of Virtual Inertia Control and Digital Protection for Microgrid Dynamic Security Considering High Renewable Energy Penetration. IET Renew. Power Gener. 2019, 13, 462–474. [Google Scholar] [CrossRef]
  5. Oliveira, W.D.; Vieira, J.P.; Bezerra, U.H.; Martins, D.A.; Rodrigues, B.d.G. Power system security assessment for multiple contingencies using multiway decision tree. Electr. Power Syst. Res. 2017, 148, 264–272. [Google Scholar] [CrossRef]
  6. Wang, J.; Song, Y.; Liu, F.; Hou, R. Analysis and application of forecasting models in wind power integration: A review of multi-step-ahead wind speed forecasting models. Renew. Sustain. Energy Rev. 2016, 60, 960–981. [Google Scholar] [CrossRef]
  7. Herrería-Alonso, S.; Suárez-González, A.; Rodríguez-Pérez, M.; Rodríguez-Rubio, R.F.; López-García, C. Efficient Wind Speed Forecasting for Resource-Constrained Sensor Devices. Sensors 2021, 21, 983. [Google Scholar] [CrossRef]
  8. Drałus, G.; Mazur, D.; Kusznier, J.; Drałus, J. Application of Artificial Intelligence Algorithms in Multilayer Perceptron and Elman Networks to Predict Photovoltaic Power Plant Generation. Energies 2023, 16, 6697. [Google Scholar] [CrossRef]
  9. Szostek, K.; Mazur, D.; Drałus, G.; Kusznier, J. Analysis of the Effectiveness of ARIMA, SARIMA, and SVR Models in Time Series Forecasting: A Case Study of Wind Farm Energy Production. Energies 2024, 17, 4803. [Google Scholar] [CrossRef]
  10. Bokde, N.D.; Tranberg, B.; Andresen, G.B. Short-term CO2 emissions forecasting based on decomposition approaches and its impact on electricity market scheduling. Appl. Energy 2021, 281, 116061. [Google Scholar] [CrossRef]
  11. Taleb, H.M.; Abu Hijleh, B. Optimizing the Power Generation of a Wind Farm in Low Wind Speed Regions. Sustainability 2021, 13, 5110. [Google Scholar] [CrossRef]
  12. García-Sánchez, T.; Mishra, A.K.; Hurtado-Pérez, E.; Puché-Panadero, R.; Fernández-Guillamón, A. A Controller for Optimum Electrical Power Extraction from a Small Grid-Interconnected Wind Turbine. Energies 2020, 13, 5809. [Google Scholar] [CrossRef]
  13. Kusznier, J.; Wojtkowski, W. IoT Solutions for Maintenance and Evaluation of Photovoltaic Systems. Energies 2021, 14, 8567. [Google Scholar] [CrossRef]
  14. Hołdyński, G.; Skibko, Z.; Borusiewicz, A. Impact of Wind Power Plant Operation on Voltage Quality Parameters—Example from Poland. Energies 2022, 15, 5573. [Google Scholar] [CrossRef]
  15. Skibko, Z.; Tymińska, M.; Romaniuk, W.; Borusiewicz, A. Impact of the Wind Turbine on the Parameters of the Electricity Supply to an Agricultural Farm. Sustainability 2021, 13, 7279. [Google Scholar] [CrossRef]
  16. Hu, Q.; Fu, L.; Ma, F.; Ji, F. Large Signal Synchronizing Instability of PLL-Based VSC Connected to Weak AC Grid. IEEE Trans. Power Syst. 2019, 34, 3220–3229. [Google Scholar] [CrossRef]
  17. Chen, Z.; Wang, X.; Kang, S. Effect of the Coupled Pitch–Yaw Motion on the Unsteady Aerodynamic Performance and Structural Response of a Floating Offshore Wind Turbine. Processes 2021, 9, 290. [Google Scholar] [CrossRef]
  18. Bandi, M.M.; Apt, J. Variability of the Wind Turbine Power Curve. Appl. Sci. 2016, 6, 262. [Google Scholar] [CrossRef]
  19. Tindal, A.; Johnson, C.; LeBlanc, M.; Harman, K.; Rareshide, E.; Graves, A. Site-specific adjustments to wind turbine power curves. In Proceedings of the AWEA Wind Power Conference, Houston, TX, USA, 1–4 June 2008. [Google Scholar]
  20. Fleming, P.; Gebraad, P.M.O.; Lee, S.; van Wingerden, J.-W.; Johnson, K.; Churchfield, M.; Michalakes, J.; Spalart, P.; Moriarty, P. Simulation comparison of wake mitigation control strategies for a two-turbine case. Wind Energy 2015, 18, 2135–2143. [Google Scholar] [CrossRef]
  21. Wagner, R.; Courtney, M.; Gottschall, J.; Lindeløw-Marsden, P. Accounting for the speed shear in wind turbine power performance measurement. Wind Energy 2011, 14, 993–1004. [Google Scholar] [CrossRef]
  22. Wagner, R.; Antoniou, I.; Pedersen, S.M.; Courtney, M.S.; Jørgensen, H.E. The influence of the wind speed profile on wind turbine performance measurements. Wind Energy 2009, 12, 348–362. [Google Scholar] [CrossRef]
  23. Choukulkar, A.; Pichugina, Y.; Clack, C.T.; Calhoun, R.; Banta, R.; Brewer, A.; Hardesty, M. A new formulation for rotor equivalent wind speed for wind resource assessment and wind power forecasting. Wind Energy 2016, 19, 1439–1452. [Google Scholar] [CrossRef]
  24. Ennis, B.L.; White, J.R.; Paquette, J.A. Wind turbine blade load characterization under yaw offset at the SWiFT facility. J. Phys. Conf. Ser. 2018, 1037, 052001. [Google Scholar] [CrossRef]
  25. White, J.; Ennis, B.E.; Herges, T.G. Estimation of Rotor Loads Due to Wake Steering. In Proceedings of the 2018 Wind Energy Symposium, Kissimmee, FL, USA, 8–12 January 2018. [Google Scholar]
  26. Lee, H.; Lee, D. Wake impact on aerodynamic characteristics of horizontal axis wind turbine under yawed flow conditions. Renew. Energy 2019, 136, 383–392. [Google Scholar] [CrossRef]
  27. Meng, Q.; He, Y.; Hussain, S.; Lu, J.; Guerrero, J.M. Day-ahead economic dispatch of wind-integrated microgrids using coordinated energy storage and hybrid demand response strategies. Sci. Rep. 2025, 15, 26579. [Google Scholar] [CrossRef]
  28. Lee, J.C.Y.; Stuart, P.; Clifton, A.; Fields, M.J.; Perr-Sauer, J.; Williams, L.; Cameron, L.; Geer, T.; Housley, P. The Power Curve Working Group’s assessment of wind turbine power performance prediction methods. Wind Energy Sci. 2020, 5, 199–213. [Google Scholar] [CrossRef]
  29. Clifton, A.; Kilcher, L.; Lundquist, J.; Fleming, P. Using machine learning to predict wind turbine power output. Environ. Res. Lett. 2013, 8, 024009. [Google Scholar] [CrossRef]
  30. Gijón, A.; Pujana Goitia, A.; Perea, E.; Molina Solana, M.; Gómez Romero, J. Prediction of wind turbines power with physics informed neural networks and evidential uncertainty quantification. arXiv 2023. [Google Scholar] [CrossRef]
  31. Zierath, J.; Rachholz, R.; Woernle, C. Field test validation of Flex5, MSC.Adams, alaska/Wind and SIMPACK for load calculations on wind turbines. Wind Energy 2016, 19, 1201–1222. [Google Scholar] [CrossRef]
  32. Lydia, M.; Kumar, S.S.; Selvakumar, A.I.; Kumar, G.E.P. A comprehensive review on wind turbine power curve modeling techniques. Renew. Sustain. Energy Rev. 2014, 30, 452–460. [Google Scholar] [CrossRef]
  33. Sohoni, V.; Gupta, S.C.; Nema, R.K. A Critical Review on Wind Turbine Power Curve Modelling Techniques and Their Applications in Wind Based Energy Systems. J. Energy 2016, 2016, 8519785. [Google Scholar] [CrossRef]
  34. Stanley, A.P.; Roberts, O.; Lopez, A.; Williams, T.; Barker, A. Turbine scale and siting considerations in wind plant layout optimization and implications for capacity density. Energy Rep. 2022, 8, 3507–3525. [Google Scholar] [CrossRef]
  35. Hu, S.-Y.; Cheng, J.-H. Performance evaluation of pairing between sites and wind turbines. Renew. Energy 2007, 32, 1934–1947. [Google Scholar] [CrossRef]
  36. Liu, Y.; Fu, Y.; Huang, L.-L.; Zhang, K. Reborn and upgrading: Optimum repowering planning for offshore wind farms. Energy Rep. 2022, 8, 5204–5214. [Google Scholar] [CrossRef]
  37. Pallabazzer, R. Parametric analysis of wind siting efficiency. J. Wind Eng. Ind. Aerodyn. 2003, 91, 1329–1352. [Google Scholar] [CrossRef]
  38. IEC 61400-12-2:2022; Wind Energy Generation Systems—Part 12-2: Power Performance of Electricity Producing Wind Turbines Based on Nacelle Anemometry. International Electrotechnical Commission (IEC): Geneva, Switzerland, 2022.
  39. Burton, T.; Jenkins, N.; Sharpe, D.; Bossanyi, E. Wind Energy Handbook; John Wiley & Sons: Chichester, UK, 2011. [Google Scholar]
  40. Shin, D.; Ko, K. Application of the Nacelle Transfer Function by a Nacelle-Mounted Light Detection and Ranging System to Wind Turbine Power Performance Measurement. Energies 2019, 12, 1087. [Google Scholar] [CrossRef]
  41. Vahidzadeh, M.; Markfort, C.D. Modified Power Curves for Prediction of Power Output of Wind Farms. Energies 2019, 12, 1805. [Google Scholar] [CrossRef]
  42. Yang, M.; Peng, T.; Zhang, W.; Su, X.; Han, C.; Fan, F. Abnormal Data Identification and Reconstruction Based on Wind Speed Characteristics. CSEE J. Power Energy Syst. 2025, 11, 612–622. [Google Scholar] [CrossRef]
  43. Woodward, A.; Nixon, W. Wind Lidars as Permanent Wind Speed Reference Systems on Operational Wind Farms. 2025. Available online: https://www.zxlidars.com/wp-content/uploads/2025/07/Informative-Guide-for-Wind-Lidars-as-Permanent-Wind-Speed-Reference-Systems-FINAL.pdf (accessed on 18 August 2025).
  44. Carbajo Fuertes, F.; Markfort, C.D.; Porté-Agel, F. Wind Turbine Wake Characterization with Nacelle-Mounted Wind Lidars for Analytical Wake Model Validation. Remote Sens. 2018, 10, 668. [Google Scholar] [CrossRef]
  45. Zou, M.; Fang, D.; Djokic, S.Z.; Di Giorgio, V.; Langella, R.; Testa, A. Evaluation of wind turbine power outputs with and without uncertainties in input wind speed and direction data. IET Renew. Power Gener. 2020, 14, 2801–2809. [Google Scholar] [CrossRef]
  46. Bilendo, F.; Meyer, A.; Badihi, H.; Lu, N.; Cambron, P.; Jiang, B. Applications and Modeling Techniques of Wind Turbine Power Curve for Wind Farms—A Review. Energies 2023, 16, 180. [Google Scholar] [CrossRef]
  47. Gill, S.; Stephen, B.; Galloway, S. Wind Turbine Condition Assessment Through Power Curve Copula Modeling. IEEE Trans. Sustain. Energy 2011, 3, 94–101. [Google Scholar] [CrossRef]
  48. Pryor, S.C.; Shepherd, T.J.; Barthelmie, R.J. Interannual variability of wind climates and wind turbine annual energy production. Wind Energy Sci. 2018, 3, 651–665. [Google Scholar] [CrossRef]
  49. Lee, J.-T.; Kim, H.-G.; Kang, Y.-H.; Kim, J.-Y. Determining the Optimized Hub Height of Wind Turbine Using the Wind Resource Map of South Korea. Energies 2019, 12, 2949. [Google Scholar] [CrossRef]
  50. Ashkenazy, Y.; Yizhaq, H. The diurnal cycle and temporal trends of surface winds. arXiv 2022, arXiv:2207.01103. [Google Scholar] [CrossRef]
  51. Niang, S.A.A.; Drame, M.S.; Gueye, A.; Sarr, A.; Toure, M.D.; Diop, D.; Ndiaye, S.O.; Talla, K. Temporal Dynamics of Energy Production at the Taïba Ndiaye Wind Farm in Senegal. Discov. Energy 2023, 3, 6. [Google Scholar] [CrossRef]
  52. Poupkou, A.; Zanis, P.; Nastos, P.; Papanastasiou, D.; Melas, D.; Tourpali, K.; Zerefos, C. Present climate trend analysis of the Etesian winds in the Aegean Sea. Theor. Appl. Climatol. 2011, 106, 459–472. [Google Scholar] [CrossRef]
  53. Laurila, T.K.; Sinclair, V.A.; Gregow, H. Climatology, variability, and trends in near-surface wind speeds over the North Atlantic and Europe during 1979–2018 based on ERA5. Int. J. Climatol. 2021, 41, 2253–2278. [Google Scholar] [CrossRef]
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Article Metrics

Citations

Article Access Statistics

Journal Statistics
Multiple requests from the same IP address are counted as one view.
Download PDF
Frequency Regulation Characteristics of Molten Salt Thermal Energy Storage-Integrated Coal-Fired Power Units
Previous
Assessing Positive Energy District Potential: A Case Study in Central Italy
Next