Page 143 - IJOCTA-15-1

An International Journal of Optimization and Control: Theories & Applications
ISSN: 2146-0957 eISSN: 2146-5703
Vol.15, No.1, pp.137-154 (2025)
https://doi.org/10.36922/ijocta.1664

RESEARCH ARTICLE

Modeling the renewable energy development in T¨urkiye with
optimization

Neyre Tekbıyık-Ersoy *

Energy Systems Engineering, Faculty of Engineering, Cyprus International University, 99258, Nicosia, North
Cyprus via Mersin 10, T¨urkiye
nersoy@ciu.edu.tr

ARTICLE INFO ABSTRACT

Article History: Due to increasing electricity consumption in T¨urkiye, energy supply had to
Received: 9 August 2024 increase significantly in the last few decades. As T¨urkiye has been highly de-
Accepted: 4 January 2025 pendent on energy imports, this had an effect on country’s economy. However,
Available Online: 27 January 2025 especially in the last two decades, T¨urkiye increased the utilization of renew-
ables. This can significantly improve the energy security and the economy of
Keywords:
the country. This paper investigates the development of renewable energy in
Modeling
Optimization T¨urkiye and aims to determine the main drivers behind the significant devel-
Renewable energy opment recorded in the last decades. In order to do that, past data have been
T¨urkiye analyzed and a constrained optimization problem has been constructed. This
optimization problem aims to minimize the Mean Absolute Error (MAE) be-
AMS Classification 2010: tween the real and estimated renewable electricity capacity levels in T¨urkiye.
90C30; 65K05; 65K10 In order to analyze the effect of increasing the number of modeling parameters,
this paper considers three cases with different number of modeling parameters.
The results in each case reveal the optimum weights and the modeling param-
eters that should be preferred for modeling the installed renewable energy
capacity in T¨urkiye. The results of the proposed approach, called Mean Abso-
lute Error based OPTimization (MAEOPT), are also compared with that of
multiple linear regression (MLR). The obtained MAE values for the best six
models reported in this paper (three for MAEOPT and three for MLR) are less
than 5%, which indicates an excellent performance. The results also indicate
the superiority of MAEOPT over MLR, in terms of MAE and Mean Absolute
Percentage Error (MAPE).

1. Introduction has strong renewable energy potential, it has poor
natural gas and petroleum reserves, and its lignite
T¨urkiye has a strong geopolitical position due coal reserves are not sufficient to meet country’s
to the fact that it borders the Mediterranean, fuel needs. As energy production in T¨urkiye is
Aegean and Black Seas. It forms a natural en- heavily based on fossil fuels, T¨urkiye depends on
ergy bridge between the major natural gas and energy imports in order to satisfy the demand.
3
oil producers in the Middle East and Caspian Sea The study in claims that T¨urkiye supplies more
regions, and the consumer markets in Europe. 1 than 50% of its energy demand from imported
T¨urkiye’s electricity consumption has increased energy sources (oil and natural gas having the
from 50.1 TWh to 272 TWh in the last 30 years largest share in importation) and that 98% of
2
(from 1990 to 2019). The energy production level the natural gas consumption is met by impor-
4
in the country also increased during the same pe- tation. Similarly, according to the study in, in
riod. However, it has not been enough to meet 2019, almost 69% of the primary energy demand
the ever-increasing demand. Although T¨urkiye
*Corresponding Author
137