The Environment and Wind Energy Production by Analyzing Noise

Filtering in Wind Signals to Improve the Efficiency of Energy

Systems

Naim Baftiu

, Ana Atanasova

, Tatjana A. Pacemska

and Petre Lameski

Faculty of Computer Science,“Goce Delcev” University, Stip, North Macedonia

Faculty of Computer Science and Engineering Science, “Ss. Cyril and Methodius” University in Skopje, North Macedonia

Keywords: Noise Filtering, Wind Signals, FIR, IIR, Wavelet Transform, Kalman Filter.

Abstract: Noise filtering is an important process for improving the accuracy and efficiency of signals captured by wind

sensors, which are used to monitor and optimize the performance of wind-based energy systems. Wind signals

often contain interference and noise, which can complicate analysis and decision-making related to energy

production and turbine maintenance. In this context, noise filtering helps improve the quality of data collected

by the sensors and allows for a more accurate assessment of wind speed and direction, contributing to more

efficient energy management. This is crucial for optimizing energy production, reducing costs, and increasing

the sustainability of energy systems. The use of signal filtering algorithms can significantly enhance the

performance of energy systems by eliminating the negative impacts of external factors, such as acoustic

pollution and interference from other sources. Noise in wind signals, caused by atmospheric disturbances,

sensor inaccuracies, and electromagnetic interference, reduces the efficiency of energy systems. This study

focuses on implementing digital filters to improve signal quality, thereby enhancing turbine performance. The

implementation of FIR, IIR, wavelet transform, Kalman filters, and spectral analysis aims to optimize wind

energy production.

1 INTRODUCTION

Wind energy has become the cornerstone of

renewable energy solutions, providing a permanent

option for fossil fuels. However, the efficiency of

wind energy systems is highly dependent on the

quality of the input signals, specifically wind speed

and direction. These signals are critical for the precise

control of turbine operations, including blade pitch

adjustment, yaw control, and optimal energy

conversion.

Unfortunately, wind signals are often corrupted

by noise stemming from multiple sources.

Atmospheric turbulence, caused by unpredictable

weather patterns, introduces random variations in the

signals. Additionally, sensor inaccuracies, resulting

https://orcid.org/0000-0001-9432-9293

https://orcid.org/0009-0004-6094-9995

https://orcid.org/0000-0002-5336-1796

from calibration errors or environmental wear, further

degrade the quality of the measurements. Moreover,

electromagnetic interference, arising from nearby

electronic devices or power lines, can significantly

distort the recorded data.

The presence of noise in wind signals not only

complicates data interpretation but also compromises

the operational efficiency of wind turbines. When

controllers rely on inaccurate or noisy signals,

turbines may operate sub- optimally, leading to

decreased energy output, increased mechanical stress,

and higher maintenance costs.

To address these challenges, advanced noise

filtering techniques have emerged as crucial tools for

enhancing the reliability and accuracy of wind signal

data. This paper investigates the application of

112

Baftiu, N., Atanasova, A., Pacemska, T. A. and Lameski, P.

The Environment and Wind Energy Production by Analyzing Noise Filtering in Wind Signals to Improve the Efﬁciency of Energy Systems.

DOI: 10.5220/0014377900004848

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 2nd International Conference on Advances in Electrical, Electronics, Energy, and Computer Sciences (ICEEECS 2025), pages 112-119

ISBN: 978-989-758-783-2

various digital filters, including Finite Impulse

Response (FIR), Infinite Impulse Response (IIR),

Wavelet Transforms, and Kalman Filters, to mitigate

noise in wind signals. Furthermore, spectral analysis

techniques are employed to identify and target

specific noise frequencies, enabling tailored filtering

solutions. By improving signal quality, these filtering

methods contribute to the optimization of turbine

performance and overall energy production. The

study highlights the potential of digital filtering in

renewable energy applications, offering a framework

for future advancements in wind energy systems. The

goal is to develop robust methodologies that ensure

reliable energy generation, even under challenging

environmental conditions.

2 MANUSCRIPT PREPARATION

Noise filtering in wind energy systems has been a

significant research topic, as accurate signal

processing is crucial for optimizing turbine

performance. Several studies have explored different

techniques for noise reduction in wind signals,

focusing on both time-domain and frequency-domain

filtering approaches. (Ackermann, 2005) discussed

the impact of wind turbulence on energy production

and the necessity of accurate wind measurements for

improved efficiency. The study highlighted that

incorrect signal processing could lead to errors in

turbine control, reducing overall energy output. In

recent years, digital filtering techniques such as FIR

and IIR have gained popularity for denoising wind

signals. (Manyonge et al. 2012) investigated the

effectiveness of FIR and IIR filters in mitigating

sensor noise in wind speed measurements. Their

results demonstrated that FIR filters maintain signal

phase integrity, making them suitable for real-time

applications, while IIR filters provide efficient noise

suppression with fewer computational resources.

Another important approach is the use of Wavelet

Transform, which has been widely applied in non-

stationary signal processing. (Liu, X., Zhang, Y.

2022), examined the role of wavelet-based filtering in

wind speed prediction, showing that wavelet

decomposition effectively isolates noise while

preserving important signal features. This method

was particularly useful in identifying transient noise

components that traditional filters struggled to

remove. Kalman filtering has also been explored for

real-time noise reduction in dynamic wind energy

systems. (Qazi et al. 2021) implemented an adaptive

Kalman filter to refine wind speed estimates,

resulting in a significant improvement in turbine

power output predictions. Kalman filters excel in

tracking changes over time, making them a powerful

tool for applications where wind conditions fluctuate

rapidly.

Apart from filtering techniques, spectral analysis

methods have been employed to understand the

frequency characteristics of wind signal noise.

(Bhardwaj et al. 2023) utilized Fast Fourier

Transform (FFT) to identify dominant noise

frequencies in wind measurements, allowing for the

design of targeted filtering solutions. Their research

demonstrated that combining spectral analysis with

adaptive filtering methods leads to superior noise

reduction performance. While these studies have

made significant contributions to the field, challenges

remain in developing filtering techniques that balance

computational efficiency with high accuracy. This

paper builds upon previous research by integrating

multiple filtering approaches—FIR, IIR, Wavelet

Transform, and Kalman Filtering—along with

spectral analysis to enhance wind signal quality. By

comparing their effectiveness, this study aims to

provide insights into the optimal filtering strategy for

wind energy applications.

3 BASIC METHODS, MODELS

FOR NOISE FILTERING

Basic methods and models for noise filtering are

techniques and algorithms used to eliminate or

minimize the impact of noise from signals to preserve

valuable information and improve data quality.

These methods are crucial in fields such as signal

processing, telecommunications, and wind energy

applications, where interference and noise can affect

the accuracy of measurements and analyses.

Here are some of the most used methods and

models for noise filtering:

FIR Filters (Finite Impulse Response)

Description: FIR filters are filters with a

finite impulse response, meaning they are

composed of a limited number of

coefficients. They are used to filter noise

without affecting the original signal.

Advantage: They are easy to design and

implement and provide full stability.

IIR Filters (Infinite Impulse Response)

Description: IIR filters have an infinite

impulse response, using an infinite number

of coefficients. These filters can be more

efficient in preserving information and are

faster compared to FIR filters.

The Environment and Wind Energy Production by Analyzing Noise Filtering in Wind Signals to Improve the Efﬁciency of Energy Systems

113

Advantage: They use fewer coefficients,

making them more efficient in terms of

resource consumption.

Wavelet Transform

Description: This transformation is used to

break down a signal into its simpler

components at different frequency levels. It

is particularly useful for filtering noise when

there are rapid or nonlinear variations in the

signal.

Advantage: It allows the signal to be

decomposed into components that can be

processed more easily and is effective for

signals that contain sudden changes.

Kalman Filters

Description: Kalman filters are advanced

algorithms for signal filtering that use

various mathematical models to predict and

correct errors in signal measurements.

Advantage: They are highly effective for

tracking and filtering signals in

environments with high noise and have been

widely used in navigation and robotic

applications.

Spectral Analysis

Description: This method involves breaking

down the signal into its frequency

components. By analyzing

the

energy

spectrum,frequencies that contain noise

can be identified and eliminated.

Advantage: It provides an efficient

method for filtering noise by removing

frequencies that fall outside the range of

interest.

Adaptive Filtering

Description: This type of filtering uses

algorithms that dynamically adjust to

changes in the signal and noise. For

example, an adaptive filter can adjust in real-

time to handle changes in noise levels.

Advantage: It can filter noise under varying

conditions where noise levels change

continuously.

Statistical Filtering

Description: This method uses the statistics

of the signal to identify and eliminate noise,

relying on probabilistic models to

distinguish the true signal from the noise.

Advantage: It is used in situations where the

noise follows a known pattern or can be

statistically evaluated.

All these methods are useful in different

circumstances and can be employed to improve the

accuracy of measurements and optimize the

performance of systems that rely on captured signals.

The choice of method depends on the nature of the

noise and the specific requirements of the application.

In the context of climate change and renewable

energy production, particularly wind energy, ensuring

the accuracy of wind data analysis is crucial for

effective planning and decision-making. Wind speed

data is often subject to various types of noise, which

can arise from sensor inaccuracies, atmospheric

turbulence, and other external factors.

To ensure that the data used for wind energy

applications is reliable, different filtering techniques

are applied to remove noise while preserving the

essential characteristics of the wind signal.

This study applies multiple filtering methods,

including Finite Impulse Response (FIR), Infinite

Impulse Response (IIR), Wavelet Transform and

Kalman Filter. Each of these methods offers unique

advantages in noise reduction and signal smoothing.

Below, we discuss the methodologies and

characteristics of each filtering approach. (Wang, J.,

Zhang, F. 2024)

The noise from turbines can negatively impact

wind energy production for several reasons:

Impact on sensors and inaccurate

measurements: Noise can affect the sensors

used to measure wind speed and direction,

leading to inaccurate data. This can result in

incorrect decisions for turbine adjustments,

reducing energy production efficiency.

Loss in production capacity: In addition to

the interference in measurements, noise can

cause an increase in wasted energy due to

vibrations and oscillations, which affect the

turbine’s performance. This may lead to

lower energy production and may even

cause mechanical damage to the turbine over

long periods.

Reduced accuracy in performance

analysis: Noise can mask the natural

variations in wind, making it difficult to

accurately assess turbine performance.

Without accurate analysis, energy

management and operational optimization

become more challenging.

Impact on energy management: Noise can

cause difficulties in forecasting energy

supply and can affect energy management

systems that rely on accurate data for

planning and optimization. This may lead to

resource wastage or increased operational

costs.

ICEEECS 2025 - International Conference on Advances in Electrical, Electronics, Energy, and Computer Sciences

114

Impact on turbine lifespan: Noise and

vibrations can cause mechanical stress on

turbine components, shortening their

operational life and increasing the need for

frequent maintenance.

For these reasons, noise filtering and improving

measurement accuracy are essential to ensure higher

efficiency in wind energy production and to enhance

the sustainability of wind energy systems.

The noise from wind turbines can negatively impact

residents living near wind turbines for several

reasons:

Impact on mental and physical health:

The constant noise and vibrations caused by

the turbines can lead to stress, anxiety,

insomnia, and fatigue for residents. Studies

have shown that prolonged exposure to

noise can contribute to mental health

issues and affect the nervous system.

Impact on quality of life: Noise can affect

the quality of life for residents by creating an

uncomfortable environment. It can interfere

with daily activities such as resting,

conversations, and outdoor activities,

leading to feelings of dissatisfaction and

tension.

Sleep problems: The noise from turbines

can impact sleep, causing issues like

insomnia and deterioration in sleep quality.

This can affect both physical and mental

health, leading to fatigue and reduced

performance during the day.

Impact on the local environment: Noise

can affect wildlife in the area surrounding

the turbines, disturbing animals and

disrupting local ecosystems. This can have a

negative impact on residents who are

connected to nature and the nearby

environment.

Social impact: Wind turbines can cause

social division within local communities.

Some residents may feel disturbed by the

noise, leading to tensions between those who

support wind energy projects and those who

are dissatisfied with the noise impact.

To manage these impacts, it is important for wind

turbine projects to be carefully planned and managed,

considering acceptable noise levels and distances

from residential areas to ensure minimal impact on

nearby residents. (P. Denholm and M. Hand, 2011)

3.1 Finite Impulse Response (FIR)

Filter

The FIR filter is a type of digital filter that applies a

finite number of coefficients to modify the wind

speed signal. This filter is widely used due to its

stability and linear phase response, making it an

effective tool for noise reduction in wind speed data.

The FIR filter does not rely on past outputs, making

it inherently stable and suitable for filtering out high-

frequency noise. In this study, we applied the FIR

filter to smooth the wind speed data, reducing

unwanted fluctuations caused by environmental

factors or sensor inaccuracies.

3.2 Infinite Impulse Response (IIR)

Filter

Unlike the FIR Filter, the IIR Filter uses both past

inputs and past outputs to generate filtered values.

This makes it more computationally efficient while

achieving the desired filtering effect with fewer

coefficients. The IIR filter can effectively reduce

noise while maintaining a balance between signal

distortion and smoothness. For this project, we

applied the IIR filter to compare its performance with

FIR filtering in terms of noise reduction and signal

preservation.

3.3 Wavelet Transform

Wavelet Transform is a powerful signal processing

technique used for decomposing a signal into

different frequency components. Unlike traditional

filters, wavelets allow for localized time- frequency

analysis, making them highly effective for detecting

and removing transient noise while preserving the

wind signal's characteristics. The application of

Wavelet Transform in this study helped in denoising

the wind speed signal by removing high- frequency

components while keeping the underlying wind

patterns intact.

3.4 Kalman Filter

The Kalman Filter is a recursive algorithm used for

estimating the true value of a noisy signal. It works

by predicting the system's future state and updating

the estimates based on new observations. This

filtering method is particularly useful for applications

where real-time data processing is required. We

applied the Kalman Filter to the wind speed dataset to

enhance the accuracy of the signal and reduce the

impact of measurement errors.

The Environment and Wind Energy Production by Analyzing Noise Filtering in Wind Signals to Improve the Efﬁciency of Energy Systems

115

4 EXPERIMENTS AND RESULTS

4.1 Analysis of Wind Noise

Components

Wind speed signals, like any real-world

measurements, are affected by noise. Noise in wind

data can distort analysis, leading to inaccurate power

predictions for wind turbines. To ensure high-quality

data, it is important to analyses the sources and

characteristics of noise and apply appropriate filtering

techniques to mitigate their effects. Noise in wind

speed data originates from various factors, including:

Sensor Noise: Missing values in the dataset

were imputed using linear interpolation for

continuous variables like wind speed and

temperature, ensuring the time series

continuity.

Atmospheric Turbulence: For categorical

variables like wind direction, a forward fill

approach was employed, where missing

values were filled with the most recent

available observation.

Obstructions and Terrain Effects: Objects

such as buildings, trees, and mountains alter

wind flow, creating localized disturbances in

the wind speed measurements.

Measurement Resolution and Sampling

Rate: If wind speed data is recorded at a very

high sampling rate, it can capture short-term

variations that may be considered noise. On

the other hand, a low sampling rate may

cause loss of important variations.

Instrumental and Transmission Errors:

Data transmission errors, power failures, and

signal dropouts can introduce gaps, outliers,

or artefacts in the recorded wind data.

Conclusion: Noise in wind speed data

originates from multiple sources, including

sensor limitations, atmospheric turbulence,

and environmental obstructions.

Understanding the characteristics of noise

allows the application of appropriate filtering

techniques to enhance data quality. By

employing methods such as FFT analysis and

filtering, we can ensure cleaner wind speed

signals, leading to more accurate predictions

and efficient wind turbine operation.

4.2 Experimental Results

In this section, we present the results obtained from

applying the noise analysis techniques outlined in the

previous chapter to the wind speed dataset. The

primary objective was to evaluate the effectiveness of

different noise reduction methods in enhancing the

quality of the wind speed signals. For this purpose,

both the raw and filtered data are analyzed, and

various statistical measures are used to quantify the

impact of noise.

4.2.1 Visualization of Results

To assess the performance of each noise reduction

model, we visualized the comparison between actual

and predicted wind energy production values. By

plotting the actual energy production against the

predicted values generated by each model, we gain

valuable insight into the accuracy and stability of the

models over time. These visualizations help us

understand how well the models capture the true

variations

wind

energy

production

and whether

they can maintain accuracy despite the presence of

noise. In the plots, we observe the degree of

alignment between the actual and predicted values for

each filtering method. A closer match indicates a

more accurate model, while greater discrepancies

suggest that the model may not effectively replicate

the true energy production trends. Additionally, by

analyzing these plots over time, we can assess the

consistency and reliability of the models under

varying wind conditions. These visualizations are

essential for determining which model provides the

best balance between reducing noise and preserving

the key features of the wind energy signal. They offer

a clear, intuitive representation of model performance

and help identify areas where further refinement

might be needed.

Raw Data Analysis: To begin, we visualize the raw

wind speed data to observe the presence of noise. As

shown in Figure 4.1, the raw data exhibits significant

fluctuations and high-frequency components that are

indicative of noise. These fluctuations are not

characteristic of true wind behavior, and they obscure

the underlying wind patterns that are crucial for

accurate analysis. (Figure 1).

Wind Speed with FIR Filter: This plot showcases

the effect of a Finite Impulse Response (FIR) filter on

wind speed data over time. The original wind speed

is represented in blue, showing a highly fluctuating

and noisy pattern. The red line represents the FIR-

filtered wind speed, which smooths the variations

while maintaining the general trend of the data. FIR

filters are known for their phase linearity, meaning

that they preserve the timing of the signal components

ICEEECS 2025 - International Conference on Advances in Electrical, Electronics, Energy, and Computer Sciences

116

Figure 1. Raw Wind Speed Data.

without distortion. The graph reveals that the FIR

filter effectively removes high-frequency noise while

retaining important patterns in the wind speed

variations. (Figure 2).

Figure 2. FIR Filtered Wind Speed Data.

Wind Speed with IIR Filter: This plot presents

the effect of an Infinite Impulse

Response (IIR)

filter on wind speed data. The blue line represents the

original, unfiltered wind speed measurements, which

exhibit significant fluctuations. The green line shows

the filtered wind speed using the IIR approach.

Compared to FIR filtering, the IIR filter provides a

smoother response but may introduce phase

distortion. The filtering effect is subtler compared to

FIR, as IIR filters tend to provide better frequency

response with fewer coefficients. This makes them

efficient for real-time applications, but the output

may lag slightly due to recursive computations.

(Figure 3).

Figure 3. IIR Filtered Wind Speed Data.

Wind Speed with Wavelet Denoising: This final

plot illustrates wind speed data denoised using

wavelet transformation. The blue line represents the

original wind speed readings, while the cyan line

corresponds to the denoised signal. Wavelet.

denoising works by decomposing the signal into

multiple frequency components and selectively

removing high-frequency noise. Unlike traditional

FIR and IIR filters, wavelet denoising is particularly

useful for signals with non-stationary characteristics,

such as wind speed, where noise and trends vary over

time. The plot suggests that this technique effectively

smooths the data while preserving the key

fluctuations in wind behavior. (Figure 4).

Figure 4. Wavelet Denoised Wind Speed Data.

Wind Speed with Kalman Filter: This figure

displays wind speed data processed through a Kalman

filter, a state-estimation algorithm that smooths noisy

measurements. The original wind speed (in blue)

exhibits strong fluctuations, while the Kalman-

filtered result (in pink) appears significantly

smoothed yet responsive to changes.

The Kalman filter dynamically adjusts its

predictions based on past values and measurement

uncertainties, making it particularly effective for

The Environment and Wind Energy Production by Analyzing Noise Filtering in Wind Signals to Improve the Efﬁciency of Energy Systems

117

tracking time- series data like wind speed. Unlike FIR

and IIR filters, which rely solely on predefined

coefficients, the Kalman filter adapts to incoming

data, reducing noise while preserving underlying

trends. (Figure 5).

Figure 5. Kalman Filtered Wind Speed Data.

5 CONCLUSION

When selecting the best filter for wind speed data

processing, several factors must be considered,

including noise reduction efficiency, computational

complexity, response time, and adaptability to real-

time changes. Each filtering method has its own

strengths and limitations, making them more suitable

for different applications. Below is a detailed

comparison of the filters based on their effectiveness

in various aspects.

The Kalman filter stands out as one of the most

effective techniques for filtering wind speed data due

to its ability to dynamically adapt to changes in the

system. Unlike traditional filters that apply fixed

coefficients to smooth out noise, the Kalman filter

continuously updates its internal model based on new

observations. This makes it particularly useful when

working with time-varying signals such as wind

speed, which is inherently unpredictable. One of its

biggest advantages is its ability to minimize noise

while preserving the true underlying signal, even in

the presence of uncertainty. This is because the

Kalman filter does not just smooth out high-

frequency fluctuations but also provides an optimal

estimation of the system's state, considering both

measurement noise and process uncertainty.

However, despite its high accuracy and adaptability,

the Kalman filter comes with a higher computational

cost compared to other filters like FIR and IIR.

It requires knowledge of the system’s noise

characteristics, which may not always be

straightforward to determine. Additionally, improper

tuning of the filter parameters can lead to inaccurate

estimations or slow convergence.

The Wavelet Transform is highly effective at

filtering wind speed data because it decomposes the

signal into different frequency components, allowing

for selective noise removal. This makes it superior to

standard frequency-domain filters, such as FIR and

IIR, when dealing with non-stationary signals—those

that have time-dependent fluctuations. Wind speed

data often contains random bursts of noise due to

turbulence, sensor interference, or environmental

disturbances. Unlike traditional filtering methods that

apply the same filtering operation across the entire

signal, wavelet-based filtering adapts to different

frequency bands, ensuring that the essential

components of the signal remain intact while

eliminating

unwanted noise. One limitation of

wavelet filtering is that it can sometimes introduce

distortions, especially in regions where the signal has

sharp transitions.

Additionally, selecting the appropriate wavelet

function and decomposition level requires expertise,

as an improper choice could lead to signal

degradation rather than improvement.

The Finite Impulse Response (FIR) filter is

widely used in signal processing due to its ability to

maintain a linear phase response, meaning that all

frequency components of the signal experience the

same time delay. This ensures that the shape of the

signal remains unaltered, making FIR filtering an

ideal choice when signal integrity is a priority. FIR

filters are designed with fixed coefficients and rely

only on past input values, making them inherently

stable and resistant to numerical errors. Additionally,

they allow for precise control over the filter response,

enabling selective attenuation of unwanted frequency

components. However, FIR filters tend to be

computationally expensive compared to IIR filters, as

they require a larger number of coefficients to achieve

the same level of noise suppression. This makes them

less suitable for real-time applications where

computational efficiency is crucial.

The Infinite Impulse Response (IIR) filter is

known for its efficiency in removing unwanted noise

while using fewer coefficients compared to an FIR

filter. This makes it a preferred choice for real-time

applications where computational resources are

limited. One of the main advantages of the IIR filter

is that it uses feedback, meaning that its output

depends not only on the current input values but also

on past outputs. This enables the filter to achieve a

stronger noise reduction effect with a lower

computational cost.

ICEEECS 2025 - International Conference on Advances in Electrical, Electronics, Energy, and Computer Sciences

118

However, this feedback mechanism introduces a

potential downside: phase distortion, which can cause

delays or shifts in certain frequency components of

the signal. Another concern with IIR filters is their

potential for instability, especially when not designed

properly. Unlike FIR filters, which are always stable,

IIR filters can become unstable if their parameters are

not carefully chosen.

The choice of filter depends on factors such as

computational constraints, the type of noise present,

and the required precision in the wind speed data

analysis.

REFERENCES

Ackermann, T. 2005 Wind Power in Power Systems. Royal

Institute of Technology Stockholm, Sweden, Wiley,

ISBN 0-470-85508-8.

Manyonge, A.W., Ochieng, R.M., Onyango, F.N.,

Shichikha, J.M. 2012. Mathematical Modelling of Wind

Turbine in a Wind Energy Conversion System: Power

Coefficient Analysis. Applied Mathematical Sciences,

vol. 6, no. 91, pp. 4527–4536.

Liu, X., Zhang, Y. 2022. Wind Speed Prediction Using

LSTM and Traditional Statistical Models. Scopus

Journal of Energy Systems, vol. 45, no. 7, pp. 993–

1005.

Bhardwaj, A., Kumar, P., Sharma, R. Spectral, 2023.

Analysis and Adaptive Filtering for Wind Signal

Processing. IEEE Transactions on Sustainable Energy,

vol. 13, no. 2, pp. 1012–1023. DOI:

10.1109/TSTE.2023.3276384.

Wang, J., Zhang, F. 2024. Comparative Study of Time- Series

Models for Wind Energy Prediction. IEEE Access, vol.

10, pp. 157890–157901. DOI:

10.1109/ACCESS.2024.3102034.

Qazi, M.A., Ahmed, S., Rehman, S. Adaptive,2021.

Kalman Filtering for Wind Speed Prediction in Energy

Systems. Journal of Renewable Energy Research, vol.

9, no. 4.

International Energy Agency (IEA), 2024. "European

Union – World Energy Investment 2024,".

[Online].Available: https://www.iea.org/reports/world-

energy- investment-2024/european-union.

P. Denholm and M. Hand,2011. "Grid flexibility and

storage required to achieve very high penetration of

variable renewable electricity,”," Energy Policy, , Vols.

vol. 39, no. 3,, p. pp. 1817–1830.

International Energy Agency (IEA), 2022. Renewable

Energy Market Update, Paris, France.

European Commission, 2021. Renewable Energy

Directive II, Brussels, Belgium.

The Environment and Wind Energy Production by Analyzing Noise Filtering in Wind Signals to Improve the Efﬁciency of Energy Systems

119