WSEAS TRANSACTIONS on POWER SYSTEMS Si̇ mge Yi̇ ği̇ t, Safi̇ ye Turgay,

DOI: 10.37394/232016.2024.19.12 Çi̇ ğdem Cebeci̇ , Esma Sedef Kara

Time-Stratified Analysis of Electricity Consumption: A Regression and

Neural Network Approach in the Context of Turkey

SİMGE YİĞİT1, SAFİYE TURGAY1, ÇİĞDEM CEBECİ2, ESMA SEDEF KARA3

1
Department of Industrial Engineering,
Sakarya University,
54187, Esentepe Campus Serdivan-Sakarya,
TURKEY
2
Department of Electrical Machinery and Material Supply,
Energy Branch Directorate,
Sakarya Municipality, Saski General Directorate,
Sakarya,
TURKEY
3
Rüstempaşa Mahallesi,
İpekyolu Caddesi, No.120,
54600, Sapanca-Sakarya,
TURKEY

Abstract: - This study aims to apply seasonality and temporal effects in the analysis of electricity consumption
in Turkey as a case mixed with regression and neural network methodologies. The study goal is to increase
knowledge about the features and trending forces behind electricity usage which provide informed
recommendations for smart energy planning and regulation. Comparing and contrasting the regression and
neural network models makes it possible to carry out a thorough analysis of the merits and demerits of each
model. Moreover, the examination of the limits of the models and their performance in forecasting electricity
consumption patterns over the long term is done. The results of this study have a significant impact on power
forecasting techniques, and they have meaningful effects on the policymakers, planners and utilities in
Turkey. Understanding the story of the use of electricity around the world is very important for the
development of sustainable energy policies, resource provision, and the maintenance of reliable and smart
energy networks in the country.

Key-Words: - Electricity Consumption, Time-Stratified Analysis, Regression Modeling, Neural Network

Approach, Energy Forecasting, Turkey, Sustainable Energy Policies, Resource Optimization.

Received: March 17, 2023. Revised: January 2, 2024. Accepted: February 15, 2024. Published: April 2, 2024.

1 Introduction Turkey's own geographical and climatic

The growing need for electricity in Turkey, being peculiarities as well as her largest share of activities,
connected with the mounting challenges of modern which determine electricity, demand structurally, all
electricity systems, requires a deep investigation are responsible for the development of a distinctive
into consumers' behavior to welfare the formulation electricity consumption profile. Knowing the
of a desirable electricity plan and the enactment of alterations in temporal attributes for instance
power policies. It is important to pay attention to seasonality, temperature, and time-of-day is very
that as time-stratified analysis of electricity important for having optimal energy infrastructures,
consumption in the Turkey case is done with and for making sure grid reliability and formulating
traditional regression models and with the advanced sustainable policies.
neural network methods. We hope to track the time- Then, as a solution to the determined
based fluctuation of electricity demand in the hope restrictions of traditional methods, the artificial
of exposing the driving factors of variation in usage neural networks, which are the tools for non-linear
upshot.

E-ISSN: 2224-350X 96 Volume 19, 2024

 WSEAS TRANSACTIONS on POWER SYSTEMS Si̇ mge Yi̇ ği̇ t, Safi̇ ye Turgay,
DOI: 10.37394/232016.2024.19.12 Çi̇ ğdem Cebeci̇ , Esma Sedef Kara

pattern building and complex relationship traffic, In the next phase, we move into the dimension
are utilized. of neural networks, taking advantage of their
Through applying a time-based model, the competence in unmasking perplexing patterns and
research is committed to observing the complexity associations within huge datasets. The neural
of electricity consumption at different intervals, network model adapts and learns from historical
seasonal factors, economic cycles, and possible patterns of consumption, which gives it high
emotional issues. Net electricity generation is accuracy for making predictions and discovering the
defined as the total of all electricity production nuances that traditional regression models never see.
minus all the same amount of electricity consumed The main variables reflecting a study are; Real
in a given region or country. The net electricity Gross Domestic Product; Population; Quantity of
consumption is significant to consider in terms of Vehicles; Foreign Trade; and Industry. Electricity
energy demand, ensure the electricity demand and demand is a complex and multi-influencing
supply balance. For operational purposes, net phenomenon, which depends on economic
electricity consumption is commonly measured in activities, population growth, the effect of weather
megawatts or gigawatts (MWh/GWh). This gauge elements, and technological developments. When
indicates the difference between the total power comprehending the volatile nature of electricity
output formed over some time (for example, every usage, it is imperative for good energy planning and
hour, day, month, or year) and the total power sustainable development.
consumption. The demand for grid electricity is This article is about showing the temporal
significant data for planners of energy, investments variations of Turkey's electricity consumption using
in the area of infrastructure, and the design of time-stratified analysis, which is a combination of
energy policy, just to mention a few. The net traditional regression analysis as well as the
electric power consumption of a portion or country utilization of modern artificial neural networks
is used to identify the variations of its energy (ANNs).
demand, plan energy outputs, and utilize energy The layout of the paper comes down to the
resources properly. Data on net electricity elements of the methodology, which in turn are
consumption was obtained from a variety of regression analysis and neural network
sources, including energy companies, government modeling. This time-layered analysis provides the
agencies, and international energy organizations. findings that are accommodative to the design and
This information is used in many areas such as formation of plans and strategies concerning energy,
electricity consumption analysis, energy demand which again, helps in policy formulation. Factors
forecasts, and energy policy development. However, that are seen to be substantial in one time. Period
data on net electricity consumption is often could change in later times and this is where
published with a lag and it is important to follow regression analysis comes to the fore. On the other
official energy market data sources or producers for hand, neural networks enable the researcher to have
up-to-date information. a deep understanding of non-linear relationships
The primary objective of this research is to between factors. Through this, predictions that are
enhance the accuracy and depth of our more accurate are achieved. The anticipated effects
understanding of electricity consumption patterns in of these findings on energy policy or planning
Turkey through a two-fold approach: firstly, include the discussion of effective approaches and
employing a traditional regression model that the prospects for more competent forecasting and
encompasses fundamental temporal variables, and sound decision-making in the volatile reality of
secondly, leveraging the power of neural networks consumption are illustrated. This study adds to the
to capture complex, non-linear relationships emerging branch in energy analytics of investigating
inherent in the data. the performance of regression models as a simple
The first phase of our analysis involves alternative to deep neural networks in electricity
constructing a comprehensive regression model. consumption prediction. As such, they bring
This model considers variables such as time of day, significant impacts for the above-mentioned actors,
day of the week, and seasonality, aiming to quantify that is, energy planners, policymakers, and utility
the impact of these factors on electricity companies, which in turn are used as important and
consumption. From historical data, the regression relevant inputs in allocating resources and
model gives an insight into the time patterns of developing sustainable energy strategies in Turkey.
consumption, which provides a background analysis As we delve into the depths of time-stratified
for later methods of more intricate mechanisms. electricity consumption analysis, the subsequent
sections of this research will unfold the

E-ISSN: 2224-350X 97 Volume 19, 2024

 WSEAS TRANSACTIONS on POWER SYSTEMS Si̇ mge Yi̇ ği̇ t, Safi̇ ye Turgay,
DOI: 10.37394/232016.2024.19.12 Çi̇ ğdem Cebeci̇ , Esma Sedef Kara

methodology, data, results, and discussions, investigations address the place of temperature and
ultimately leading to a nuanced understanding of the or weather in total energy consumption. In the spirit
temporal intricacies shaping Turkey's electricity of the diversity of weather in Turkey, these
landscape. correlations are of significant value to the accuracy
The remainder of this paper is structured as of the modeling and forecasting, [26], [27], [28],
follows: Section 2 includes an overview of the [29], [30].
pertinent literature. Section 3 defines the model It is with this basic objective that the current
definition and formulation, encompassing the research will bring contributions by building upon
mathematical model of the time-stratified analysis existing knowledge and making a definitional
of electricity consumption. Section 4 structures a addition to Electricity Consumption Analysis, which
case study, summarizing the key results of our is valid in the Turkish context.
proposed approach in comparison to the current
state of affairs. Finally, in Section 5, we present our
concluding remarks. 3 Methodology
Herewith the given analysis is undertaken which
focuses on electricity consumption in Turkey.
2 Literature Survey Grasping a holistic view by using both traditional
The literature survey explores existing research regression analysis and advanced artificial neural
related to time-stratified analysis of electricity networks (ANNs). The main purpose of the research
consumption, with a focus on regression and neural is to find hidden relationships and identify changes
network approaches, within the specific context of that occur inside the time domain of electricity
Turkey. Some of the studies cover the temporal demand. This kind of information is very valuable
analysis of electricity consumption which as it tries to support the future policymakers, as well
emphasizes the importance of considering temporal as all energy planners, and of course their
dynamics for accurate forecasting and efficient stakeholders.
energy planning, [1], [2], [3], [4], [5]. Some of the Regressions analysis is very stable. Through an
studies highlight the ability of neural networks to ordered review of historical data, key variables such
capture non-linear relationships and intricate as consumption of energy with time can be easily
patterns, contributing to more accurate predictions identified. Then, besides the stem learning
in electricity consumption modeling, [6], [7], [8], algorithms like regression and correlation analysis,
[9], [10]. Some of the researchers have explored the which are excellent in terms of capturing any non-
influence of geographical and socioeconomic linearity, the artificial neural networks are applied to
factors on electricity consumption, [11], [12]. Given improve the forecasting accuracy.
Turkey's unique characteristics, understanding how The research uses a data set that covers a particular
these factors interact with temporal dynamics period, which is then divided further into numerous
becomes crucial for tailored energy policies. Some periods to ensure seasonality, trend, and anomaly
of these works shed light on the country's energy injection of electricity consumption patterns being
landscape, providing valuable insights into captured in totality. Thus, by the integration of the
consumption patterns and trends that inform the regression and neural network techniques, this
present research, [13], [14], [15]. Comparative research seeks a detailed examination of time-
analyses between regression models and neural dependent features of Turkey's electricity
networks in the context of electricity consumption demand. The end purpose is to increase our
are scarce, [16], [17], [18], [19]. The authors knowledge of how the different parameters
discussing the implications of electricity influence the electricity consumption in Turkey
consumption patterns on energy policy offer a during different periods. By using the strengths of
broader perspective, [20], [21], [22]. The present both interpretability associated with regression
research aims to contribute to this discourse by analysis and the flexibility provided by neural
providing insights specifically tailored to the networks, this methodological dual approach allows
Turkish context. Advancements in time series us to gain a deeper understanding of demand
analysis have been crucial for refining response and how different factors affect it in the
methodologies in electricity consumption studies. temporal sense. Finally, the study is considered as a
The integration of advanced artificial intelligence in useful tool for decision-makers in the energy sector,
the current technology is an effort made to improve and it will help prepare effective performance plans
the accuracy of timeline forecasts and deepen and permanent actions that are used by energy
understanding of it, [23], [24], [25]. The other management planners in Turkey.

E-ISSN: 2224-350X 98 Volume 19, 2024

 WSEAS TRANSACTIONS on POWER SYSTEMS Si̇ mge Yi̇ ği̇ t, Safi̇ ye Turgay,
DOI: 10.37394/232016.2024.19.12 Çi̇ ğdem Cebeci̇ , Esma Sedef Kara

In electricity consumption time stratified analysis, For the neural network, it could be the same
regression and neural network approaches on the MSE or another suitable loss function.
one hand provide a flexible and integrating
methodology. The following paragraph is about the 3.1.4 Optimization
regression model, as well as the neural network. Optimization algorithms, such as gradient descent,
are used to adjust the parameters (β for regression,
3.1 Mathematical Model 𝑊 and b for neural network) to minimize the loss:
Interpretability
3.1.1 Regression Model - Regression Model
The regression model represented as a linear o The coefficients (β) provide insights
equation: into the impact of each temporal
Y=0+1X1+2X2+…+nXn+ variable on electricity consumption.
- Neural Network
Where o Techniques like SHAP values or
Y is the dependent variable (electricity layer-wise relevance propagation
consumption) can be employed for
0 is the intercept. interpretability, attributing
1+2+…+n are the coefficients for the predictions to input features.
independent variables
X1+X2+…+Xn, representing the time of day, day of In practice, the complexity of neural network
the week, etc.) architectures can vary based on the problem's
 is the error term. intricacy. This simplified model representation
provides a foundation for understanding the core
For example, considering time of day (TOD), mathematical concepts involved in both regression
day of the week (DOW), and seasonal factors and neural network approaches.
(SEASON), the regression equation might be Policymakers can use temporal insights to
Consumtion=0+TOD.TOD+DOW.DOW+SEASON.DOSEAS implement time-specific energy-saving measures,
ON+ incentivizing consumers to reduce consumption
during peak periods. The study's findings guide
3.1.2 Neural Network Model infrastructure planning, helping to design systems
Let's consider a simple feedforward neural network: capable of handling peak demand periods
𝑌̂ = 𝑓(𝑊2 . 𝑓(𝑊1 . 𝑋 + 𝑏1 ) + 𝑏2 ) efficiently. Incorporating neural network predictions
into existing forecasting models can enhance the
Where accuracy of electricity consumption predictions,
𝑌̂ is the predicted electricity consumption. supporting more effective energy planning.
X is the input vector containing features such as The study's findings are contingent on the
time of day, day of the week, etc. quality and availability of historical data. Improved
f() is the activation function (e.g., ReLU for hidden data collection methods could enhance the accuracy
layers, linear for output layer). of predictions. While the neural network
𝑊1 and 𝑊2 are weight matrices. demonstrated superior performance, its complexity
𝑏1 and 𝑏2 are bias vectors. may pose challenges for interpretation. Future
research could focus on developing hybrid models
For instance, with a single hidden layer, the for improved interpretability. External factors such
equation might look like: as economic changes or policy shifts were not
̂
𝐶𝑜𝑛𝑠𝑢𝑚𝑝𝑡𝑖𝑜𝑛 = 𝑓(𝑊𝑜𝑢𝑡 . 𝑓(𝑊ℎ𝑖𝑑𝑑𝑒𝑛 . 𝑋 + 𝑏ℎ𝑖𝑑𝑑𝑒𝑛 ) explicitly considered. Future studies could explore
+ 𝑏𝑜𝑢𝑡 ) the integration of external variables for a more
holistic analysis.
3.1.3 Training
Both models involve training by minimizing a loss
function. For the regression model, it might be 4 Case Study
Mean Squared Error (MSE): In this study, we conducted a time-stratified analysis
2 of electricity consumption in Turkey, employing
𝑀𝑆𝐸 = 𝑁1 ∑ 𝑁(𝑌̂𝑖 − 𝑌𝑖 )
𝑖=1 both regression and neural network approaches. The
research aimed to enhance our understanding of

E-ISSN: 2224-350X 99 Volume 19, 2024

 WSEAS TRANSACTIONS on POWER SYSTEMS Si̇ mge Yi̇ ği̇ t, Safi̇ ye Turgay,
DOI: 10.37394/232016.2024.19.12 Çi̇ ğdem Cebeci̇ , Esma Sedef Kara

temporal patterns and drivers behind electricity calculated. The data were analyzed with Google
consumption, providing valuable insights for energy Colab.
planning and policy formulation (Figure 1). By systematically implementing this
methodology, the study aims to provide a
comprehensive understanding of time-stratified
electricity consumption in Turkey, comparing the
efficacy of traditional regression models with
advanced neural network approaches.
The proposed algorithm presents a hybrid model
to capture the temporal dynamics of electricity
consumption in Turkey by combining regression
Fig. 1: The proposed model outline and neural network approaches. Through extensive
analysis and model comparison, the algorithm aims
The regression model revealed significant to provide accurate forecasts and valuable insights
relationships between temporal variables (time of for sustainable energy planning and policy
day, day of the week, and seasonality) and formulation.
electricity consumption. Interpretation of
coefficients highlighted the impact of specific time- 1. Obtaining and organizing data
related factors on consumption patterns. The neural Data on the amount of electricity consumption and
network outperformed the regression model in variables were found in bibliographies containing
capturing complex, non-linear relationships within and analyzing various statistical documentation.
the electricity consumption dataset. The model These data are organized in an Excel file and
demonstrated adaptability to intricate patterns, converted into tables (Figure 2).
providing more accurate predictions than the
traditional regression approach.
Comparative analysis highlighted the strengths
and limitations of each approach. The regression
model offered interpretability, while the neural
network excelled in capturing intricate patterns.
Both models demonstrated generalizability,
effectively predicting electricity consumption trends
in unseen data. The neural network, however,
showcased superior adaptability to diverse patterns.
Insights from this study have direct implications for
energy policy in Turkey. Understanding temporal
dynamics is crucial for optimizing resource Fig. 2: Data set sample
allocation, ensuring grid reliability, and developing
sustainable energy strategies. Turkey's electricity 2. Analyzing the data
consumption between 1975 and 2021 and the The data were analyzed through Google Colab file,
independent variables [ (Population, Gross missing data detection, data categorization,
Domestic Product, Number of Vehicles, Foreign correlations, and their effects on consumption
Trade ($), Industry (TL)] that are thought to affect amounts were analyzed (in Figure 3, Figure 4,
this consumption amount are used. The study Figure 5, Figure 6, Figure 7).
proceeds in line with two objectives. In the first
objective, the effects of the independent variables on 3. Selection of the appropriate solution method
the net consumption amount are analyzed and it is Since the problem is aimed at forecasting,
examined at how many times which variable affects regression, and artificial neural network models
it. Secondly, it is aimed to determine whether these were selected among the models that could be
methods are successful in predicting net electricity suitable for this.
consumption and to calculate which method is more
successful with performance measures after
prediction by regression analysis and artificial
neural networks method. Prediction values were
compared with actual values and error metrics were

E-ISSN: 2224-350X 100 Volume 19, 2024

 WSEAS TRANSACTIONS on POWER SYSTEMS Si̇ mge Yi̇ ği̇ t, Safi̇ ye Turgay,
DOI: 10.37394/232016.2024.19.12 Çi̇ ğdem Cebeci̇ , Esma Sedef Kara

Fig. 3: The relationship between Net Consumption

and Real GDP Fig. 7: The Relationship between Net Consumption
and Industry

4. Regression and ANN modeling

The problem was modeled using the necessary
coding on Google Colab (Table 1).

Table 1. Prediction results and regression results

Fig. 4: The relationship between Net Consumption

and Population

5. Evaluation of performances
Fig. 5: The relationship between Net Consumption For the performance values of the models, their
and Vehicles scores were analyzed and the mean squared error
and the margin of error and the differences between
them were found in Figure 8 and Table 2.

Fig. 6: The relationship between Net Consumption

and Foreign Trade

Fig. 8: Statistical Results

E-ISSN: 2224-350X 101 Volume 19, 2024

 WSEAS TRANSACTIONS on POWER SYSTEMS Si̇ mge Yi̇ ği̇ t, Safi̇ ye Turgay,
DOI: 10.37394/232016.2024.19.12 Çi̇ ğdem Cebeci̇ , Esma Sedef Kara

Table 2. Performance analysis results of regression and neural network results in an even
more accurate and better-performing performing the
hybrid-forecasting model. The coefficient value (α)
optimized for harmony to take from both
procedures, thus, reducing prediction error and
making the model more comprehensive.
The result obtained from this analysis just can
contribute to a more precise forecasting of the
amount of electricity, that people consume in
Turkey. Thus, energy turbines and officials can
make use of these forecasts to plan well resources
As a result, the regression model was found to be based on the use and development of physical
more successful. infrastructures. Knowledge of seasonality in
electricity demand enables purposeful strategy in a
place where high seasonal demand is addressed and
5 Conclusion conserving off-seasons to make. The time-step
In this study, we conducted a time-stratified analysis modeling method allows for identifying and
of electricity consumption in Turkey using both anticipating the incidence of irregularities in
regression analysis and artificial neural network consumption, which is crucial for increasing the
(ANN) models. In this context, the goal was to show adaptability and resilience of the energy system.
the different trends in electricity consumption and In summary, this study reveals informative
think of possible power planning strategies and knowledge about the regional, seasonal, and
ideas for new energy policies. Coming together organizational factors involving electricity
prediction techniques by using regression and neural consumption in Turkey.
network approaches, a model hybrid is improved for The set of regression analysis, neural network
achieving more effectiveness in forecasting and modeling, and hybrid approaches is not only an
capturing electricity consumption dynamics effective model but also a very robust tool that is
complexity. well-suited for understanding and forecasting the
In the case of the Mean Squared Error, it dynamic behavior of energy demand over time. This
concluded that the model of regression was ideal for is linked as well with the inspiration of policies that
this set of data. The network of Artificial Neural recognize the use of renewable energies in Turkey.
Network (ANN) is designed in such a way to be The benefit will ultimately be a sustainable and
able to analyze the most complicated data sets that resilient energy future for the country.
are not analyzed by regression analysis. Acknowledgement:
In the analysis, it very complicated periodical flow It is an optional section where the authors may write
of power use that had changing nature of the time a short text on what should be acknowledged
and difference of the periods, respectively. regarding their manuscript.
Seasonal phenomena, economic difficulties as
well as activity during a specific period had quite a
special influence on the electricity References:
demand. Regression Models presented various [1] Nsangou, J.C., Kenfack, J., Nzotcha, U.,
factors concerning electricity consumption and the Ekam, P.S.N., Voufo, J., Tamo, T.T.,
economic impacts on selected Explaining household electricity consumption
determinants. However, economic indicators, using quantile regression, decision tree and
population growth, and climate factors were at artificial neural network, Energy, Vol. 250, 1
different degrees of significance in regard to July 2022, 123856.
consumption changes at various times. The [2] Schneider, N., Strielkowski, W., Modelling
networks of neurons with their abilities to show the unit root properties of electricity data—A
non-linear correlations were telling us something general note on time-domain applications,
hidden among the complexity of data. They Physica A: Statistical Mechanics and its
practiced their mastery of not only technical but also Applications, Vol. 618, 15 May 2023,
macroeconomic feedback, especially throughout 128685.
periods of non-linear growth in consumption [3] Sharafi, S., Kazemi, A., Amiri, Z., Estimating
rates. Sometimes, combining the prognostic strength energy consumption and GHG emissions in
crop production: A machine learning

E-ISSN: 2224-350X 102 Volume 19, 2024

 WSEAS TRANSACTIONS on POWER SYSTEMS Si̇ mge Yi̇ ği̇ t, Safi̇ ye Turgay,
DOI: 10.37394/232016.2024.19.12 Çi̇ ğdem Cebeci̇ , Esma Sedef Kara

approach, Journal of Cleaner Production, Vol. [13] Marriette Sakah, M., Can, S.R., Diawuo, F.A.,
408, 1 July 2023, 137242. Sedzro, M.D., Kuhn, C., A study of appliance
[4] Bacanin, N., Jovanovic, L., Zivkovic, M., ownership and electricity consumption
Kandasamy, V., Antonijevic, M., Deveci, M., determinants in urban Ghanaian households,
Strumberger, I., Multivariate energy Sustainable Cities and Society, Vol.
forecasting via metaheuristic tuned long-short 44, January 2019, Pages 559-581.
term memory and gated recurrent unit neural [14] Saryazdi, S.M.E., Etemad, A., Shafaat, A.,
networks, Information Sciences, Vol. 642, Bahman, A.M., A comprehensive review and
September 2023, 119122. sensitivity analysis of the factors affecting the
[5] Kesriklioğlu, E., Oktay, E., Karaaslan, A., performance of buildings equipped with
Predicting total household energy Variable Refrigerant Flow system in Middle
expenditures using ensemble learning East climates, Renewable and Sustainable
methods, Energy, Vol. 276, 1 August 2023, Energy Reviews, Vol. 191, March 2024,
127581. 114131.
[6] Michael P. B. , Abrasaldo, S. J. Z., Andreas [15] Zuin, G., Buechler, R., Sun, T., Zanocco, C.,
W. Kempa-Liehr, A.W., A systematic review Galuppo, F., Veloso, A., Rajagopal, R.,
of data analytics applications in above-ground Extreme event counterfactual analysis of
geothermal energy operations, Renewable and electricity consumption in Brazil: Historical
Sustainable Energy Reviews, Vol. 189, Part impacts and future outlook under climate
B, January 2024, 113998. change, Energy, Vol. 281, 15 October 2023,
[7] Mehmood, U. M., Chun, D., Shan, Z., 128101.
Hyunjoo Han, H., Jeon, G., Chen, K., A [16] Lazzari, F., Mor, G., Cipriano, J., Gabaldon,
review of the applications of artificial E., Grillone, B., Chemisana, D., Solsona, F.,
intelligence and big data to buildings for User behaviour models to forecast electricity
energy-efficiency and a comfortable indoor consumption of residential customers based
living environment, Energy and Buildings, on smart metering data, Energy Reports, Vol.
Vol. 202, 1 November 2019, 109383. 8, November 2022, Pages 3680-3691.
[8] Tso, G.K.F., Yau, K.K.W., Predicting [17] Lahmar, S., Maalmi, M., Idchabani, R.,
electricity energy consumption: A comparison Investigating adaptive sampling strategies for
of regression analysis, decision tree and optimal building energy performance using
neural networks, Energy, Vol. 32, Issue artificial neural networks and kriging
9, September 2007, Pages 1761-1768. surrogate models, Journal of Building
[9] Oreshkin, B.N., Dudek, G., Pełka, P., Turkina, Engineering, Vol. 82, 1 April 2024, 108341
R., N-BEATS neural network for mid-term [18] Moustafa, M., Ruifeng, T., Wen, J., Bo, W.,
electricity load forecasting, Applied Energy, Ullah, A., Mohamad, H.A.E., Cheng, H.,
Vol. 293, 1 July 2021, 116918. Modeling of wavy water film by application
[10] Civak, H., Küren, C., Turgay, S., Examining of artificial neural network - a state of art
the effects of COVID-19 Data with Panel study, Nuclear Engineering and Design, Vol.
Data Analysis, Social Medicine and Health 417, February 2024, 112731.
Management (2021) Vol. 2: 1-16 Clausius [19] Zou, Y., Lin, Z., Li, D., Liu, Z.C.,
Scientific Press, Canada, DOI: Advancements in Artificial Neural Networks
10.23977/socmhm.2021.020101 ISSN 2616- for health management of energy storage
2210. lithium-ion batteries: A comprehensive
[11] Kheiri, F., A review on optimization methods review, Journal of Energy Storage, Vol. 73,
applied in energy-efficient building geometry Part C, 15 December 2023, 109069.
and envelope design, Renewable and [20] Palaniappan, S., Karuppannan, S., Velusamy,
Sustainable Energy Reviews, Vol. D., Categorization of Indian residential
92, September 2018, Pages 897-920. consumers electrical energy consumption
[12] Shine, P., Murphy, M.D., Upton, J., Scully, pattern using clustering and classification
T., Machine-learning algorithms for techniques, Energy, Vol. 289, 15 February
predicting on-farm direct water and electricity 2024, 129992.
consumption on pasture based dairy farms, [21] Lee, D., Ooka, R., Ikeda, S., Choi, W., Kwak,
Computers and Electronics in Agriculture, Y., Model predictive control of building
Vol. 150, July 2018, Pages 74-87. energy systems with thermal energy storage in
response to occupancy variations and time-

E-ISSN: 2224-350X 103 Volume 19, 2024

 WSEAS TRANSACTIONS on POWER SYSTEMS Si̇ mge Yi̇ ği̇ t, Safi̇ ye Turgay,
DOI: 10.37394/232016.2024.19.12 Çi̇ ğdem Cebeci̇ , Esma Sedef Kara

variant electricity prices, Energy and [30] Nsangou, J.C.., Kenfack, J., Nzotcha, U.,
Buildings, Vol. 225, 15 October 2020, Ekam, P.S:N., Voufo, J., Tamo, T.T.,
110291. Explaining household electricity consumption
[22] Mohapatra, S.K., Mishra,S., Tripathy, H.K., using quantile regression, decision tree and
Alkhayyat, A., A sustainable data-driven artificial neural network, Energy, Vol.
energy consumption assessment model for 250,2022,123856.
building infrastructures in resource constraint
environment, Sustainable Energy
Technologies and Assessments, Vol. 53, Part Contribution of Individual Authors to the
C, October 2022, 102697 Creation of a Scientific Article (Ghostwriting
[23] Ghadami, N., Gheibi, M., Kian, Z., Faramarz, Policy)
M.G., Naghedi, R., Eftekhari, M., Fathollahi- The authors equally contributed to the present
Fard, A.M., Dulebenets, M.A., Tian, G., research, at all stages from the formulation of the
Implementation of solar energy in smart cities problem to the final findings and solution.
using an integration of artificial neural
network, photovoltaic system and classical Sources of Funding for Research Presented in a
Delphi methods, Sustainable Cities and Scientific Article or Scientific Article Itself
Society, Vol. 74, November 2021, 103149 No funding was received for conducting this study.
[24] Sefeedpari, P., Rafiee, S., Akram, A., Chau,
K., Pishgar-Komleh, S.H., Prophesying egg Conflict of Interest
production based on energy consumption The authors have no conflicts of interest to declare.
using multi-layered adaptive neural fuzzy
inference system approach, Computers and Creative Commons Attribution License 4.0
Electronics in Agriculture, Vol. (Attribution 4.0 International, CC BY 4.0)
131, December 2016, Pages 10-19 This article is published under the terms of the
[25] Kayali, S., Turgay, S., Predictive Analytics Creative Commons Attribution License 4.0
for Stock and Demand Balance Using Deep https://creativecommons.org/licenses/by/4.0/deed.en
Q-Learning Algorithm. Data and Knowledge _US
Engineering (2023) Vol. 1: 1-10. DOI:
http://dx.doi.org/10.23977/datake.2023.01010
1.
[26] Ren, S., Hu, W., Bradbury, K., Harrison-
Atlas, D., Valeri, L.M., Murray, B., Malof,
J.M., Automated Extraction of Energy
Systems Information from Remotely Sensed
Data: A Review and Analysis, Applied
Energy, Vol. 326, 15 November 2022, 119876
[27] Blaga, R., Sabadus, A., Stefu, N., Dughir, C.,
Paulescu, M., Badescu, V., A current
perspective on the accuracy of incoming solar
energy forecasting, Progress in Energy and
Combustion Science, Vol. 70, January 2019,
Pages 119-144
[28] Hong, Y.Y., Paulo C.L., Rioflorido, P.,
Zhang, W., Hybrid deep learning and
quantum-inspired neural network for day-
ahead spatiotemporal wind speed forecasting,
Expert Systems with Applications, Vol. 241, 1
May 2024, 122645
[29] Taşkın, H., Kubat, C., Topal, B., Turgay, S.,
Comparison Between OR/Opt Techniques
and Int. Methods in Manufacturing Systems
Modelling with Fuzzy Logic International
Journal of Intelligent Manufacturing, 15, 517-
526 (2004).

E-ISSN: 2224-350X 104 Volume 19, 2024