Prediction of the Growth of Renewable Energies in the European Union

using Time Series Analysis

HOLGER KRAENZLE1, MAXIMILIAN RAMPP1, DANIEL WERNER1, JÜRGEN SEITZ1,

NEHA SHARMA2

1Duale Hochschule Baden-Württemberg,

Wirtschaftsinformatik, Heidenheim 89518,

GERMANY

2AI.Cloud, Tata Consultancy Services,

INDIA

Abstract: - The whole world is affected by climate change and renewable energy plays an important role in

combating climate change. To add to the existing precarious situation, the current political events such as the

war in Ukraine mean that fossil raw materials such as oil and gas are becoming more and more expensive in the

raw material markets. This paper presents the current state of renewable energies in Germany and Europe.

Using data from the past 56 years, the predictive models ARIMA and Prophet are used to find out if the

conversion to renewable energies and the elimination of fossil raw materials in the energy sector can be

achieved in the EU. The results are compared with the target of the EU in 2030 and a long-term outlook until

2050 will be provided.

Key-Words: - Renewable energy, climate change, fossil fuels, wind energy, hydropower, solar energy,

bioenergy, Logistic Regression, time series forecasting, ARIMA, Prophet.

Received: July 8, 2022. Revised: August 24, 2023. Accepted: September 28, 2023. Published: November 20, 2023.

1 Introduction

Climate change is the long-term changes in climate

patterns of the Earth, which has wide-ranging and

potentially catastrophic impacts on the environment.

We can observe shifts in weather patterns, rising

global temperatures, the increase in extreme weather

events rising sea levels, melting ice caps and

glaciers, more frequent and severe heatwaves,

droughts, floods, and disruptions to ecosystems. It

also poses risks to human health, food security, and

economies. The primary driver of contemporary

climate change is the increase in greenhouse gas

emissions, primarily carbon dioxide (CO2), methane

(CH4), and nitrous oxide (N2O), from human

activities, such as burning fossil fuels (coal, oil, and

natural gas), deforestation, and industrial processes,

[1], [2], [3], [4].

Climate change is a huge challenge of the 21st

Century and the most important action item is to

reduce greenhouse gas emissions from the energy

sector. A transition from conventional energy

sources like oil, gas, and coal to renewable energies

like solar-, hydro-, bio-, or wind energy is required.

Renewable energy is energy that is derived from

sources that are naturally replenished and are

considered environmentally sustainable. These

sources include solar energy, wind energy,

hydropower, geothermal energy, and biomass.

These sources produce little to no greenhouse gas

emissions during electricity generation, making

them a crucial part of efforts to combat climate

change. They also reduce air and water pollution,

decrease dependence on fossil fuels, and create jobs

in the renewable energy industry. Renewable energy

technologies include solar panels (photovoltaic),

wind turbines, hydroelectric dams, geothermal

power plants, and bioenergy facilities, [5], [6], [7],

[8].

Integrating renewable energy into the energy

mix involves building infrastructure, such as solar

and wind farms, and developing energy storage

solutions to address intermittency issues (i.e., the

variability of renewable energy sources). The

United Nations knows that and therefore declared

the seventh sustainable development goal:

affordable and clean energy, [9].

The European Union (EU) also recognized the

necessity and set goals to increase the share of

renewable energy in total energy consumption.

According to the EU, the energy sector is

responsible for more than 75% of the emissions in

the EU and therefore it is the most important field to

WSEAS TRANSACTIONS on COMPUTERS

DOI: 10.37394/23205.2023.22.26

Holger Kraenzle, Maximilian Rampp,

Daniel Werner, Jürgen Seitz, Neha Sharma

E-ISSN: 2224-2872

225

Volume 22, 2023

take action. For 2020 the EU target was to a share of

renewable energy consumption of at least 20%.

With 22% according to Eurostat’s renewable energy

statistics that goal was reached, [10]. Now for 2030,

the target is to increase the share of renewable

energy consumption to at least 32%, [11]. This

target will be compared with the prediction models

in this paper. The EU's long-term strategy for 2050

is an economy with net-zero greenhouse gas

emissions, [12]. There is no renewable energy target

for 2050 but a huge growth in renewable energy

production is necessary to reach the long-term goal.

The historical data can be analyzed and predict how

the renewable energy share could develop in the

long term.

This paper is arranged by adhering to the

standard structure. The following section focuses on

the related work done by different authors; section 3

provides information about the dataset and

algorithms used in this research; the experimental

results are presented in section 4 and section 5

covers the conclusions. Finally, the references are

mentioned in the last section.

2 Review of Related Literature

The journey from fossil fuels to renewable energy is

a long and costly process that cannot happen

overnight. Nevertheless, it is becoming more and

more important with increasingly dwindling

resources. Renewable also called regenerative

energies are collective terms. They can, as the name

reflects, regenerate themselves independently and

within a human time scale, [13]. This property

represents an irreplaceable security for the

sustainable energy supply. There are different forms

of renewable energy sources such as solar and wind

energy, hydropower, geothermal energy, and

biomass, [2], [14].

According to a study by the Federal Statistical

Office, in Q1 2022, 47.1% of the total energy

requirement in Germany could already be fed into

the power grid from renewable energies, [15]. At

30.1%, wind power is now one of the most

important renewable energies, [15]. Wind power

achieved its breakthrough in 1973 after government

subsidies made it economically attractive, [16]. A

distinction is made between onshore and offshore

wind energy. Onshore refers to the wind turbines on

land, offshore to those that are off the coast at sea.

Wind turbines are used exclusively for electricity

production and use the so-called lift principle for

this purpose, which means that the wind flowing

past sets the rotor blades in lift and thus in rotation.

This rotating motion then generates electricity.

At 5.4%, biogas is one of the third most

important renewable energies, [15]. A biogas plant

has the primary purpose of generating electricity.

For this purpose, biogas is produced in the biogas

plant, which drives a motor to generate electricity.

The biogas produced is a mixture of biomethane and

carbon dioxide in particular. Strictly speaking, the

designation bioenergy plant or bioelectricity plant is

therefore more appropriate. Nevertheless, the

production process is generally associated with the

term biogas plant, [17]. It is generated in three

overarching steps. First, the biological raw materials

are stored then the raw materials are fed into the

fermentation tanks, in which the fermentation

process that releases the biogas takes place. Finally,

the gas is converted into electricity or fed directly

into it. In particular, the decomposition process and

the generation of biogas is a natural process that

also takes place in bogs, liquid manure pits, or, in

particular, in the rumen of ruminants, [17].

In contrast to the two previously presented

technologies, photovoltaics can not only be used to

generate electricity. Solar energy can be used, for

example, by means of solar collectors to generate

heat. With the help of water vapor, it generates

electricity within solar thermal power plants, or it is

used to generate direct current through photovoltaic

systems. However, solar radiation is subject to daily,

seasonal, and regional fluctuations, [18]. This puts

solar energy in second place with 6.3%, [15].

3 Material and Method

This section highlights the dataset used and the

algorithms used to build various models in the

present research work.

3.1 Materials Used

The data used for the underlying analysis is

Renewable Energy dataset available as open data at

“Our World in Data”, [19]. There are a total of 17

tables in this dataset, and in most cases, data are

available for the years 1965 to 2021. Occasionally,

tables also contain only smaller time periods and

show different data on the expansion and use of

renewable energies around the world. For example,

the generation of energy from wind, solar, hydro,

geothermal, or biofuel energy is shown. The data is

grouped by country or continent and by year. Also,

data from the whole world can be retrieved because

data are also grouped under the attribute world.

The tables "modern-renewable-prod.csv",

“modern-renewable-energy-consumption” and

"renewable-share-energy.csv" are used for the

WSEAS TRANSACTIONS on COMPUTERS

DOI: 10.37394/23205.2023.22.26

Holger Kraenzle, Maximilian Rampp,

Daniel Werner, Jürgen Seitz, Neha Sharma

E-ISSN: 2224-2872

226

Volume 22, 2023

analysis. The table "modern-renewable-prod.csv"

contains the energy produced in terawatt hours,

divided among the different countries or continents

per year. The single record here contains the energy

production of the different types of renewable

energy in terawatt-hours. The table “modern-

renewable- energy-consumption” includes the

consumed renewable energy in terawatt hours by

energy type and by country/continent. The table

"renewable-share-energy.csv" shows the share of

renewable energies in the total energy consumption

of the country. Here, too, a grouping per year and

country or continent takes place.

3.2 Methods Used

To master the data processing, Python is used.

Python is a programming language that has diverse

application areas, [1], [2], [20]. With Python,

machine learning, automation, or even web design

can be realized. The project was realized in Google

Colaboratory using the jupyter notebook for doing

Python programming. All the necessary libraries

like numpy, pandas, matplotlib, etc. were imported

and necessary datasets were input for carrying out

the exploratory analysis using line charts, pie charts,

bar charts and scatter plots.

Several algorithms for conducting the predictive

analytics were applied to data. These include, for

example, Linear Regression, ARIMA, Exponential

Smoothing, or LSTM. In this paper, the

Autoregressive Integrated Moving Average

(ARIMA) algorithm and the automatic forecasting

procedure Prophet by Facebook are used, [21]. A

detailed explanation of ARIMA and Prophet follows

in the rest of the paper.

The Python modules that have been used to

build the model are pandas, numpy, matplotlib, os,

sklearn, statsmodel, pmdarima, math, pystan, and

fbprophet.

4 Experiments and Results

There are four different categories and degrees of

complexity for analytics as shown in Figure 1. The

first one is descriptive analytics which can be used

to analyze the data of the past and tell what

happened. The second one is diagnostic analytics

which goes deeper to find out why something

happened. A further step is predictive analytics

which uses various models to find out trends and

patterns to predict what will happen in the future.

The last stage is prescriptive analytics. It is the most

complex one and here the tool would tell based on

the results of predictive analytics what someone

should do in the future to reach a goal, [22], [23].

In this paper, the focus lies on descriptive and

predictive analytics. To discover the chosen dataset

some descriptive analyses will be done first. The

dataset was prepared for analysis and to help derive

various insights. The original data file was

downloaded to a database (Microsoft SQL Server

2017 Database) which made it easier to work as the

original data was converted to a simple text file.

Fig. 1: Different analytics categories

4.1 Descriptive Analytics

First, the dataset is explored and different analyses

with the German and EU data are carried out. It

must be considered that the data of the EU contains

the data of Germany.

Figure 2 shows the TWh generated within

Germany. The years are shown on the X-axis and it

ranges from the year 1965 to the year 2021. The Y-

axis shows the generated energy in TWh. Over the

years, the generated energy from water energy was

mostly around 20 TWh. Sometimes it was less than

20 TWh, but sometimes even more than 20 TWh.

Wind energy was the most promoted in Germany.

Here, slightly less than 120 TWh was produced in

2021. In comparison, solar and other renewables

produced only about 50 TWh. Wind power is very

strongly developed in Germany. The reason for this

is that Germany has a very large wind power area

due to its proximity to the Alps in the south and the

North Sea and Baltic Sea in the north, [24], [25].

WSEAS TRANSACTIONS on COMPUTERS

DOI: 10.37394/23205.2023.22.26

Holger Kraenzle, Maximilian Rampp,

Daniel Werner, Jürgen Seitz, Neha Sharma

E-ISSN: 2224-2872

227

Volume 22, 2023

Fig. 2: Line chart of the German development

Within the EU, the picture is different for hydro

energy as depicted in Figure 3. The production of

renewable energies has increased here. Whereas in

1965 this was just over 200 TWh, in 2021 the

amount is just under 350 TWh. Wind, solar, and

other renewable energies show a similar picture as

in the EU. Wind energy accounts for the largest

share of renewables here, at just under 400 TWh.

The solar and other renewable energies are, as in

Germany, similarly high. When looking at the

graph, it must of course also be noted that Germany

has an influence on the graph on the development

within the EU, as Germany is also part of the EU.

Fig. 3: Line chart of the development in the EU

Furthermore, Germany and Europe can be

compared based on the produced renewable energy

categories in the last year as presented in Figure 4.

Germany produced 233.21 TWh of renewable

energy in 2021. Half of that was produced by wind

energy. The other half comes from solar energy

(21%), hydro energy (7%), and other energy

categories including bioenergy (22%).

Fig. 4: Pie chart of the German production in 2021

Figure 5 shows that the EU with all 27 member

states produced 1069 TWh of renewable energy in

2021. 36% of the produced energy came from wind.

Compared to Germany a much larger amount

originates from hydro energy. It has a share of 32%

or 343.88 TWh in total. Solar energy is 15% and

other sources are responsible for 17%.

Fig. 5: Pie chart of the EU production in 2021

Figure 6 presents that Germany was the biggest

producer of renewable energy in the EU in 2021

with 233.21 TWh. 21.82% of the total renewable

energy production is provided by Germany. The

next largest producers are Spain (124.2 TWh),

France (120.46 TWh), Italy (115.85 TWh) and

Sweden (115.19 TWh).

Fig. 6: Horizontal bar chart with the top 10

producers in the EU

Until now, only the production of renewable

energy has been analyzed. The “Our World in Data”

dataset also has data about renewable energy

consumption. The production can be compared with

the consumption in all 27 EU states. The most recent

data are available for the year 2020. Figure 7 shows

that all states consume almost exactly the amount of

renewable energy they produce. As figured out

earlier, Germany will be the biggest producer and

consumer of renewable energy in 2020.

WSEAS TRANSACTIONS on COMPUTERS

DOI: 10.37394/23205.2023.22.26

Holger Kraenzle, Maximilian Rampp,

Daniel Werner, Jürgen Seitz, Neha Sharma

E-ISSN: 2224-2872

228

Volume 22, 2023

Fig. 7: Scatter Plot of the production and

consumption of renewable energy

After a closer look at the different types of

energy production and the producing countries, the

share of renewable energy in total energy

consumption can be considered. More precisely, the

development of the share in the last decades from

1965 to 2020 in the EU will be analyzed. As shown

in Figure 8, the data shows that the share started

with a value of 6.4% in 1965 and dropped in the

years afterward. It moved sideways until 2004.

Since 2004 the percentage of renewable energy

increased a lot and rose to almost 18% in 2020

according to the dataset.

Fig. 8: Line chart of the renewable energy share in

the EU

It must be mentioned that the dataset from “Our

World in Data” doesn’t match completely with the

data cited in the Introduction, where it is said, that

the EU has reached a share of 22% of renewable

energy in 2020. It cannot be clearly stated which

data are correct but the predictions will be done with

the “Our World in Data” dataset.

4.2 Linear Regression Model

To get a first impression of how the future values

could look a simple linear regression can be done. A

linear regression ‘is used to predict the value of a

variable based on the value of another variable, [1],

[2]. Figure 9 shows that the results of the linear

regression are not very useful because the expected

share in 2050 is still lower than the share in 2020

already was. The linear regression is too simple and

not suitable for that case. More sophisticated models

need to be implemented.

Fig. 9: Linear regression of the renewable energy

share EU

4.3 ARIMA Model

One model for time series forecasting is ARIMA. It

is a statistical model for time series forecasting (cf.

Alam, 2022). It can be used for non-stationary time

series, which means data with a steadily rising line

like the renewable energy development of the EU.

ARIMA stands for Autoregressive Integrated

Moving Average and firstly uses ‘differencing to

convert a non-stationary time series into a stationary

one’ (Alam, 2022). Then the future values are

predicted with correlations and moving averages

from the historical data. The model has three

parameters: p, d, q



p is the order of the autoregressive model

(AR).



d is the degree of differencing. It shows

the number of differencing that is

necessary to get a stationary time series.



q is the order of the moving average model

(MA)

To find out the best parameters for the model,

the auto_arima function from the module prima is

used. It calculates the best model for the present

data.

Fig. 10: Calculation of the best ARIMA model

parameters

The calculation of the best ARIMA model

parameters is presented in Figure 10. Similarly, the

best model has the parameters: ARIMA (0, 2, 1)

WSEAS TRANSACTIONS on COMPUTERS

DOI: 10.37394/23205.2023.22.26

Holger Kraenzle, Maximilian Rampp,

Daniel Werner, Jürgen Seitz, Neha Sharma

E-ISSN: 2224-2872

229

Volume 22, 2023



p: 0. The order for the autoregressive

model is 0.



d: 2. Second-order differencing is required

for this dataset.



q: 1. The order of the moving average is 1.

In this paper, the model is used to predict the

share of renewable energy in the EU until 2050.

Figure 11 shows the result. The finished model

predicts a value of 26.91% in 2030, 35.87% in 2040,

and 44.83% in 2050. Compared with the EU target

of at least 32% in 2030 the ARIMA model predicts

that the goal won’t be reached with the recent

development. Net-zero greenhouse gas emissions in

2050 will neither be achieved with this development.

Fig. 11: Predicted development with ARIMA

model

4.4 Prophet Model

Prophet is a forecasting procedure by Facebook. It

can handle trends in the data as well as seasonality

and holiday effects. The module is fast, fully

automatic, and available in Python and R, [1]. The

ARIMA model is used to forecast the development

of renewable energy share in the total consumption

until 2050. With the

Prophet.make_future_dataframe function the

prediction can be done.

The following percentages of the renewable

energy share are predicted with Prophet:



2030: 23.10%





2040: 29.71%





2050: 36.21%



The lower and upper bound of the forecast is

also drawn in the diagram. For example, for 2050,

the lower bound of the forecast is 30.95% and the

upper bound is 41.39%. This is the range in which

the model estimates the value. The following line

chart in Figure 12 shows the prediction.

Fig. 12: Predicted development with FB Prophet

The Prophet model predicts lower values than

the ARIMA model. The EU target of 32% is not

reached and the forecast for 2050 is also lower.

5 Conclusion

Climate change and renewable energy are intimately

connected topics that play a crucial role in

addressing one of the most pressing global

challenges of our time: mitigating climate change

and reducing greenhouse gas emissions. Climate

change is the biggest threat to the planet Earth as it

has major impacts on humans and the environment.

Emissions from the burning of fossil fuels play an

important role in driving climate change which

should give everyone a clear goal to expand

renewable energy. Therefore, the main focus of the

study is on the time series forecast and the analysis

of the development of renewable energies in

Europe. As the forecasts show, the political goals

and target of a 32% share of renewables in total

energy consumption by 2030 are tolerable. The

ARIMA model predicts 26.91% coverage by 2030

with equal efforts. This value is 5.09% below the

target value. Unfortunately, the analysis using the

Prophet is not better, but even worse. A coverage of

only 23.10% is predicted by 2030.

From the study, it can be summarized that the

current investments and efforts to expand renewable

energies are far from sufficient to achieve the short-

term goal of 2030. But the even worse consequence

is that if these goals are not achieved, the actual

long-term goal of climate neutrality will be pushed

further and further into the background.

The future work areas and research directions

would be policy and regulatory analysis and

technology assessment. The idea is to examine the

existing policies, regulations, and international

agreements related to renewable energy adoption

and fossil fuel phase-out, as well as evaluate their

effectiveness and identify potential barriers and

opportunities for alignment with the UN timeline. In

WSEAS TRANSACTIONS on COMPUTERS

DOI: 10.37394/23205.2023.22.26

Holger Kraenzle, Maximilian Rampp,

Daniel Werner, Jürgen Seitz, Neha Sharma

E-ISSN: 2224-2872

230

Volume 22, 2023

the future, there is a need to assess the state of

renewable energy technologies, their scalability, and

the potential for innovations in areas like solar,

wind, hydropower, and energy storage, and

investigate the technological challenges and

opportunities in achieving a complete transition.

References:

[1] Sharma N., Ghosh S., Saha M. (2021). Open

Data for Sustainable Community. Advances in

Sustainability Science and Technology.

Springer, Singapore.

https://doi.org/10.1007/978-981-33-4312-

2_10.

[2] Sharma, N., De, P.K. (2023). Towards Net-

Zero Targets: Usage of Data Science for

Long-Term Sustainability Pathways.

Advances in Sustainability Science and

Technology. Springer, Singapore.

https://doi.org/10.1007/978-981-19-5244-9.

[3] K. U. R. Khan, G. H. Bisen, P. K. De and N.

Sharma, "Multivariate Regression Analysis of

Investments by Banks in Fossil Fuel Sectors

to Predict Climate Change Consequences,"

2021 IEEE 2nd International Conference on

Technology, Engineering, Management for

Societal impact using Marketing,

Entrepreneurship and Talent (TEMSMET),

2021, pp. 1-6, doi:

10.1109/TEMSMET53515.2021.9768677.

[4] Abhay Despande, Tanmay Belsare, Neha

Sharma, Prithwis De, “Univariate Time Series

Forecasting of Indian Agriculture Emissions”,

In: Singh, M., Tyagi, V., Gupta, P.K., Flusser,

J., Ören, T. (eds) Advances in Computing and

Data Sciences. ICACDS 2022.

Communications in Computer and

Information Science, vol.1614. Springer,

Cham. https://doi.org/10.1007/978-3-031-

12641-3_28.

[5] Chadha, A.S., Shinde, Y., Sharma, N., De,

P.K. (2023). Predicting CO2 Emissions by

Vehicles Using Machine Learning. In:

Goswami, S., Barara, I.S., Goje, A., Mohan,

C., Bruckstein, A.M. (eds) Data Management,

Analytics and Innovation. ICDMAI 2022.

Lecture Notes on Data Engineering and

Communications Technologies, vol.137.

Springer, Singapore.

https://doi.org/10.1007/978-981-19-2600-

6_14.

[6] Sood, A., Oza, H., Sharma, N., De, P.K.

(2023). Impact of Energy Sector on Climate

Change in India Using Forecasting Models.

In: Goswami, S., Barara, I.S., Goje, A.,

Mohan, C., Bruckstein, A.M. (eds) Data

Management, Analytics and Innovation.

ICDMAI 2022. Lecture Notes on Data

Engineering and Communications

Technologies, vol.137. Springer, Singapore.

https://doi.org/10.1007/978-981-19-2600-

6_16.

[7] Khandelwal, S., Sharma, N., De, P.K. (2023).

Real Time Carbon Emissions Calculator for

Personal Computers. In: Goswami, S., Barara,

I.S., Goje, A., Mohan, C., Bruckstein, A.M.

(eds) Data Management, Analytics and

Innovation. ICDMAI 2022. Lecture Notes on

Data Engineering and Communications

Technologies, vol.137. Springer, Singapore.

https://doi.org/10.1007/978-981-19-2600-

6_46.

[8] Belsare, T., Deshpande, A., Sharma, N. and

De, P. (2023). Application of k-Means

Method for Finding Varying Groups of

Primary Energy Household Emissions in the

Indian States. In Optimization Techniques in

Engineering (eds A. Khosla, P. Chatterjee, I.

Ali and D.

Joshi). https://doi.org/10.1002/978111990639

1.ch24.

[9] United Nations (2022, August 25). THE 17

GOALS | Sustainable Development, [Online].

https://sdgs.un.org/goals (Accessed Date:

November 16, 2023).

[10] Eurostat (2022, January 19). EU overachieves

2020 renewable energy target. Eurostat,

[Online].

https://ec.europa.eu/eurostat/web/products-

eurostat-news/-/ddn- 20220119-1 (Accessed

Date: November 16, 2023).

[11] European Commission (2022b, August 25).

Renewable energy targets, [Online].

https://energy.ec.europa.eu/index_en

(Accessed Date: November 16, 2023).

[12] European Commission (2022a, August 25).

2050 long-term strategy, [Online].

https://climate.ec.europa.eu/index_en

(Accessed Date: November 16, 2023).

[13] Schabbach T., Wesselak, V. (2012). Energy:

the future will be renewable. Technology in

focus data facts backgrounds. Berlin

Heidelberg: Springer Vieweg, p. 68.

[14] Reich G., & Reppich M. (2018). Renewable

energy technology: Overview of selected

technologies for sustainable energy supply

(2nd edition). Textbook. Wiesbaden: Springer

Vieweg. https://doi.org/10.1007/978-3-658-

20608-6.

WSEAS TRANSACTIONS on COMPUTERS

DOI: 10.37394/23205.2023.22.26

Holger Kraenzle, Maximilian Rampp,

Daniel Werner, Jürgen Seitz, Neha Sharma

E-ISSN: 2224-2872

231

Volume 22, 2023

[15] Federal Statistical Office (Destatis), Press

release no. 233 dated June 8, 2022, [Online].

https://www.destatis.de/DE/Presse/Pressemitt

eilungen/2022/06/PD22_233_43312.html

(Accessed Date: November 16, 2023).

[16] Karim ME, Karim R, Islam MT, Muhammad-

Sukki F, Bani NA, Muhtazaruddin MN.

Renewable Energy for Sustainable Growth

and Development: An Evaluation of Law and

Policy of Bangladesh. Sustainability. 2019,

vol. 11(20), 5774.

https://doi.org/10.3390/su11205774

[17] Iotti, Mattia & Bonazzi, Giuseppe. (2016).

Assessment of Biogas Plant Firms by

Application of Annual Accounts and

Financial Data Analysis Approach. Energies.

vol.9, 713, doi: 10.3390/en9090713.

[18] Shaikh, Mohd Rizwan & Shaikh, Sirajuddin

& Waghmare, Santosh & Labade, Suvarna &

Tekale, Anil. (2017). A Review Paper on

Electricity Generation from Solar Energy.

International Journal for Research in Applied

Science and Engineering Technology,

vol.887. doi: 10.22214/ijraset.2017.9272.

[19] Hannah Ritchie, Max Roser and Pablo Rosado

(2023) - “Energy” Published online at

OurWorldInData.org, [Online].

https://ourworldindata.org/energy (Accessed

Date: November 16, 2023).

[20] Raschka S, Patterson J, Nolet C. Machine

Learning in Python: Main Developments and

Technology Trends in Data Science, Machine

Learning, and Artificial

Intelligence. Information. 2020, vol.11(4),

193. https://doi.org/10.3390/info11040193.

[21] Zunic E., Korjenić K., Hodžić K., Donko D.,

(2020). Application of Facebook's Prophet

Algorithm for Successful Sales Forecasting

Based on Real-world Data. International

Journal of Computer Science and Information

Technology. vol.12, No.2, April,

https://doi.org/10.48550/arXiv.2005.07575.

[22] Bektemyssova, G., Ahmad, A. R.,

Mirzakulova, S., & Ibraeva, Z. (2022). Time

Series Forecasting by the Arima

Method. Scientific Journal of Astana IT

University, vol.11(11), pp.14–23.

https://doi.org/10.37943/HFCH4395.

[23] Abdul-Jabbar, Safa & Farhan, Alaa. (2022).

Data Analytics and Techniques: A Review.

ARO-The Scientific Journal of Koya

University, vol.10. pp.45-55, doi:

10.14500/aro.10975.

[24] Boettcher, Marita & Hoffmann, Peter &

Lenhart, Hermann & Schlünzen, Heinke &

Schoetter, Robert. (2015). Influence of large

offshore wind farms on North German

climate. Meteorologische Zeitschrift, vol.24,

pp.465-480, doi:10.1127/metz/2015/0652.

[25] Solarin SA, Bello MO. Wind energy and

sustainable electricity generation: evidence

from Germany. Environment, Development

and Sustainability. 2022, vol. 24(7), pp.9185-

9198. doi: 10.1007/s10668-021-01818-x.

Epub 2021 Sep 15. PMID: 34539230;

PMCID: PMC8441949.

Contribution of Individual Authors to the

Creation of a Scientific Article (Ghostwriting

Policy)

- Neha Sharma and Jürgen Seitz, have contemplated

the problem statement, as well as guided and

monitored the research through out its journey

- Holger Kraenzle, Maximilian Rampp and Daniel

Werner have together carried out the following

tasks:

1. Searching for the dataset

2. Carring out the preprocessing task

3. Did the exploratory data analysis

4. Build the models

Sources of Funding for Research Presented in a

Scientific Article or Scientific Article Itself

No funding was received for conducting this study.

Conflict of Interest

The authors have no conflicts of interest to declare.

Creative Commons Attribution License 4.0

(Attribution 4.0 International, CC BY 4.0)

This article is published under the terms of the

Creative Commons Attribution License 4.0

https://creativecommons.org/licenses/by/4.0/deed.en

_US

WSEAS TRANSACTIONS on COMPUTERS

DOI: 10.37394/23205.2023.22.26

Holger Kraenzle, Maximilian Rampp,

Daniel Werner, Jürgen Seitz, Neha Sharma

E-ISSN: 2224-2872

232

Volume 22, 2023