Ugur Akyol
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Ugur Akyol has a total of 25 co-authors. Here are the top co-authors by number of shared publications:
- Kamil Kahveci
Trakya University
4 shared publications Ahmet Cihan
Department of Mechanical Engineering, Faculty of Engineering and Architecture, Beykent University, Istanbul, Turkey
3 shared publications- Dinçer Akal
Department of Mechanical Engineering, Faculty of Engineering, Trakya University, Edirne, Turkey
3 shared publications - Ahmet Durak
Hayrabolu Vocational School, Tekirdağ Namık Kemal University, Hayrabolu, Tekirdağ, Turkey
2 shared publications D Akal
Mechanical Engineering Department, "Trakya University", Edirne, Turkey
2 shared publicationsK Uzuneanu
Thermal Systems and Environmental Engineering Department, "Dunarea de Jos" University of Galati, Galati, Romania
2 shared publications
Latest Publications
- Open Access
- 14 May 2026
- Published by Elsevier in SSRN Electronic Journal
Deep Learning and Explainable Artificial Intelligence for Accurate Short-Term Photovoltaic Energy Forecasting
This study investigates the predictive performance and interpretability of a Long Short-Term Memory (LSTM) model in comparison with conventional machine learning approaches, namely Extreme Gradient Boosting (XGBoost) and Random Forest (RF), for short-term photovoltaic (PV) energy forecasting. Using a high-resolution time-series dataset comprising 9,854 observations from a single PV system, the study addresses the challenge of accurately modeling highly stochastic solar energy generation while enhancing model transparency. The results demonstrate that LSTM outperforms the benchmark models, achieving the highest accuracy (R² = 0.65, RMSE ≈ 2.00, MAE ≈ 1.04), whereas XGBoost (R² ≈ 0.58) and RF (R² ≈ 0.41) exhibit comparatively lower performance, highlighting the limitations of tree-based methods in capturing temporal dependencies. Model interpretability is examined using SHapley Additive exPlanations (SHAP), which identify global solar irradiance and trigonometric temporal features (day_sin, hour_cos) as the dominant drivers of PV output. Furthermore, SHAP interaction analysis reveals a physically consistent relationship between irradiance and temperature, indicating that the positive contribution of solar radiation becomes more pronounced under elevated temperature conditions. The findings demonstrate that integrating deep learning with explainable artificial intelligence provides not only improved predictive accuracy but also enhanced transparency, addressing a critical barrier to the practical deployment of data-driven PV forecasting models.
- 14 May 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessDeep Learning and Explainable Artificial Intelligence for Accurate Short-Term Photovoltaic Energy Forecasting
This study investigates the predictive performance and interpretability of a Long Short-Term Memory (LSTM) model in comparison with conventional machine learning approaches, namely Extreme Gradient Boosting (XGBoost) and Random Forest (RF), for short-term photovoltaic (PV) energy forecasting. Using a high-resolution time-series dataset comprising 9,854 observations from a single PV system, the study addresses the challenge of accurately modeling highly stochastic solar energy generation while enhancing model transparency. The results demonstrate that LSTM outperforms the benchmark models, achieving the highest accuracy (R² = 0.65, RMSE ≈ 2.00, MAE ≈ 1.04), whereas XGBoost (R² ≈ 0.58) and RF (R² ≈ 0.41) exhibit comparatively lower performance, highlighting the limitations of tree-based methods in capturing temporal dependencies. Model interpretability is examined using SHapley Additive exPlanations (SHAP), which identify global solar irradiance and trigonometric temporal features (day_sin, hour_cos) as the dominant drivers of PV output. Furthermore, SHAP interaction analysis reveals a physically consistent relationship between irradiance and temperature, indicating that the positive contribution of solar radiation becomes more pronounced under elevated temperature conditions. The findings demonstrate that integrating deep learning with explainable artificial intelligence provides not only improved predictive accuracy but also enhanced transparency, addressing a critical barrier to the practical deployment of data-driven PV forecasting models.
- Open Access
- 21 April 2026
- Published by Elsevier in SSRN Electronic Journal
Experimental Energy and Exergy Analysis of East-Facing Bifacial PV Panels with ANN-Based Performance Prediction under High-Albedo Conditions
This study presents an experimental investigation comparing bifacial photovoltaic (bPV) and monofacial photovoltaic (mPV) panels under real field conditions in Tekirdağ, Türkiye. The bPV modules were installed at two different heights (0.5 m and 1 m) with east orientation, while the mPV module faced south. A high-reflectivity ground surface (albedo ≈ 0.80) and seasonal tilt angle optimization (30°, 40°, and 50°) were applied during the six-month experimental period.The results show that the bPV systems generated 670.28 kWh and 680.23 kWh at heights of 0.5 m and 1 m, respectively, corresponding to energy gains of 8.6% and 10.2% compared to the mPV system (617.12 kWh). Although the 1 m configuration performed slightly better, the small difference (1.5%) suggests that lower installation heights may be more cost-effective. Energy and exergy analyses indicated that bPV panels perform better during morning hours due to rear-side irradiance, while mPV panels reach peak performance at midday. Increased panel temperature reduced exergy efficiency, particularly for bifacial systems.An artificial neural network (ANN) model was developed to predict bPV power output, achieving high accuracy (R² = 0.96–0.98). Overall, the findings provide practical insights for optimizing bifacial PV system design under real operating conditions.
- 21 April 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessExperimental Energy and Exergy Analysis of East-Facing Bifacial PV Panels with ANN-Based Performance Prediction under High-Albedo Conditions
This study presents an experimental investigation comparing bifacial photovoltaic (bPV) and monofacial photovoltaic (mPV) panels under real field conditions in Tekirdağ, Türkiye. The bPV modules were installed at two different heights (0.5 m and 1 m) with east orientation, while the mPV module faced south. A high-reflectivity ground surface (albedo ≈ 0.80) and seasonal tilt angle optimization (30°, 40°, and 50°) were applied during the six-month experimental period.The results show that the bPV systems generated 670.28 kWh and 680.23 kWh at heights of 0.5 m and 1 m, respectively, corresponding to energy gains of 8.6% and 10.2% compared to the mPV system (617.12 kWh). Although the 1 m configuration performed slightly better, the small difference (1.5%) suggests that lower installation heights may be more cost-effective. Energy and exergy analyses indicated that bPV panels perform better during morning hours due to rear-side irradiance, while mPV panels reach peak performance at midday. Increased panel temperature reduced exergy efficiency, particularly for bifacial systems.An artificial neural network (ANN) model was developed to predict bPV power output, achieving high accuracy (R² = 0.96–0.98). Overall, the findings provide practical insights for optimizing bifacial PV system design under real operating conditions.
- 31 May 2025
- Published by Elsevier in International Journal of Hydrogen Energy
- Vol. 141, 88-98
Investigation of the emission values of butane, methane and methane-hydrogen mixtures used in household stove burners
- 31 May 2025
- Published by Elsevier in International Journal of Hydrogen Energy
- Vol. 141, 88-98
Investigation of the emission values of butane, methane and methane-hydrogen mixtures used in household stove burners
- Open Access
- 31 December 2024
- Published by Namik Kemal University in European Journal of Engineering and Applied Sciences
- Vol. 7 (2) , 150-158
A Water Absorption Chamber Design for Cable Capacitance Measurement Tests
XLPO cables used in ships should be tested for water absorption. Test standards require the cables be submerged in a water absorption chamber and kept at 75 0C for 14 days. Once every 7 days, the cables within the chamber are tested by applying high voltage, and their capacitance, dissipation factor, stability factor, and relative Permittivity are measured. In this study, a water absorption chamber used in the experiments is designed using fiberglass material and controlled with an ENDA ET 4420 temperature controller. It is experimentally shown that the water absorption chamber performs well.
- 31 December 2024
- Published by Namik Kemal University in European Journal of Engineering and Applied Sciences
- Vol. 7 (2) , 150-158
Open AccessA Water Absorption Chamber Design for Cable Capacitance Measurement Tests
XLPO cables used in ships should be tested for water absorption. Test standards require the cables be submerged in a water absorption chamber and kept at 75 0C for 14 days. Once every 7 days, the cables within the chamber are tested by applying high voltage, and their capacitance, dissipation factor, stability factor, and relative Permittivity are measured. In this study, a water absorption chamber used in the experiments is designed using fiberglass material and controlled with an ENDA ET 4420 temperature controller. It is experimentally shown that the water absorption chamber performs well.
- Open Access
- 21 December 2023
- Published by Namik Kemal University in European Journal of Engineering and Applied Sciences
- Vol. 6 (2) , 111-119
Synthesis and Characterization of Chitosan Based Hydrogels
Biyomedikal uygulamalarda pek çok kullanımı bulunan yumuşak doku genişleticiler cerrahi uygulamalarda karşımıza çıkmaktadır. Dental implantların osseointegrasyonunun sağlanmasında yumuşak doku genişleticiler önemli bir rol oynamaktadır. Bu çalışmada, kitosan (CS), hyaluronik asit (HA) ve jelatin (GEL) doğal polimerleri kullanılmıştır. CS, CS-HA, CS-GEL, GEL-HA ve CS-HA-GEL hidrojellerinin sentezi gerçekleştirilmiştir. HAP ile hidrojel sentezinde gama radyasyon kullanılmıştır. Fourier Dönüşümlü Kızılötesi Spektroskopisi (FTIR) ve Taramalı Elektron Mikroskobu (SEM) ile karakterizasyonları yapılmıştır. Dinamik şişme testleri distile suda gerçekleştirilmiştir. Hidrojellerin şişme davranışları ve şişme kinetikleri belirlenmiştir. Çapraz bağlayıcı konsantrasyonunun ve farklı çapraz bağlayıcıların (glutaraldehit (GA) ve hidroksiapatit (HAP)) şişme davranışına etkisi değerlendirilmiştir. Hidrojellerin difüzyonunda Fickian mekanizması görülmüştür. Sonuçlar, en yüksek kütlece şişmeye ve en uygun şişme profiline sahip hidrofilik yapılı CS-HA-1 hidrojelinin biyomedikal uygulamalarda yumuşak doku genişletici olarak kullanılabileceği göstermektedir..
- 21 December 2023
- Published by Namik Kemal University in European Journal of Engineering and Applied Sciences
- Vol. 6 (2) , 111-119
Open AccessSynthesis and Characterization of Chitosan Based Hydrogels
Biyomedikal uygulamalarda pek çok kullanımı bulunan yumuşak doku genişleticiler cerrahi uygulamalarda karşımıza çıkmaktadır. Dental implantların osseointegrasyonunun sağlanmasında yumuşak doku genişleticiler önemli bir rol oynamaktadır. Bu çalışmada, kitosan (CS), hyaluronik asit (HA) ve jelatin (GEL) doğal polimerleri kullanılmıştır. CS, CS-HA, CS-GEL, GEL-HA ve CS-HA-GEL hidrojellerinin sentezi gerçekleştirilmiştir. HAP ile hidrojel sentezinde gama radyasyon kullanılmıştır. Fourier Dönüşümlü Kızılötesi Spektroskopisi (FTIR) ve Taramalı Elektron Mikroskobu (SEM) ile karakterizasyonları yapılmıştır. Dinamik şişme testleri distile suda gerçekleştirilmiştir. Hidrojellerin şişme davranışları ve şişme kinetikleri belirlenmiştir. Çapraz bağlayıcı konsantrasyonunun ve farklı çapraz bağlayıcıların (glutaraldehit (GA) ve hidroksiapatit (HAP)) şişme davranışına etkisi değerlendirilmiştir. Hidrojellerin difüzyonunda Fickian mekanizması görülmüştür. Sonuçlar, en yüksek kütlece şişmeye ve en uygun şişme profiline sahip hidrofilik yapılı CS-HA-1 hidrojelinin biyomedikal uygulamalarda yumuşak doku genişletici olarak kullanılabileceği göstermektedir..
- 31 May 2021
- Published by Taylor & Francis in Energy Sources, Part A: Recovery, Utilization, and Environmental Effects
- Vol. 45 (3) , 8438-8457
Estimation of power output and thermodynamic analysis of standard and finned photovoltaic panels
This study deals with the estimation of power output and also thermodynamic analysis of two different photovoltaic panels. One of the panels is a standard photovoltaic module without fins (SPV), and the other one is a photovoltaic module with fins (FPV). First, a multi-layer feed-forward neural network structure is designed to estimate the daily power produced by photovoltaic modules. Furthermore, energy and exergy analyses were carried out to compare the performance of SPV and FPV panels. According to the thermodynamic analysis results using the experimental data obtained for two days (July 3, 2020 and August 4, 2020), it was calculated that the energy efficiency increased by a maximum of 8.77% and the exergy efficiency increased by a maximum of 25.9% in the FPV panel compared to the SPV panel. Moreover, considering the data obtained for each day during three months (July, August, and September), the total energy production increase in the FPV panel is approximately 6.7% compared to the SPV panel.
- 31 May 2021
- Published by Taylor & Francis in Energy Sources, Part A: Recovery, Utilization, and Environmental Effects
- Vol. 45 (3) , 8438-8457
Estimation of power output and thermodynamic analysis of standard and finned photovoltaic panels
This study deals with the estimation of power output and also thermodynamic analysis of two different photovoltaic panels. One of the panels is a standard photovoltaic module without fins (SPV), and the other one is a photovoltaic module with fins (FPV). First, a multi-layer feed-forward neural network structure is designed to estimate the daily power produced by photovoltaic modules. Furthermore, energy and exergy analyses were carried out to compare the performance of SPV and FPV panels. According to the thermodynamic analysis results using the experimental data obtained for two days (July 3, 2020 and August 4, 2020), it was calculated that the energy efficiency increased by a maximum of 8.77% and the exergy efficiency increased by a maximum of 25.9% in the FPV panel compared to the SPV panel. Moreover, considering the data obtained for each day during three months (July, August, and September), the total energy production increase in the FPV panel is approximately 6.7% compared to the SPV panel.
- 2 April 2021
- Published by Taylor & Francis in The Journal of the Textile Institute
- Vol. 113 (5) , 769-778
A mathematical model for through-air drying process of yarn bobbins
In this study, a mathematical model has been developed to simulate the through-air drying process of yarn bobbins. For this purpose, experimental data was obtained in a prototype experimental set up by passing pressurized hot air through the wool yarn bobbins. First of all, the physical phenomenon expressing the drying process has been reduced to the heat transfer problem and then a mathematical model has been written for the drying process which also includes the convective term. Using the experimental data, the coefficient included in the mathematical model was found by the extremal method. So, an inverse problem was solved. The accuracy of the model was checked by comparing with the experimentally obtained temperature values after solving a direct heat transfer problem in the given conditions. Good correlation between the obtained model results and the experimental results shows the accuracy of the mathematical model.
- 2 April 2021
- Published by Taylor & Francis in The Journal of the Textile Institute
- Vol. 113 (5) , 769-778
A mathematical model for through-air drying process of yarn bobbins
In this study, a mathematical model has been developed to simulate the through-air drying process of yarn bobbins. For this purpose, experimental data was obtained in a prototype experimental set up by passing pressurized hot air through the wool yarn bobbins. First of all, the physical phenomenon expressing the drying process has been reduced to the heat transfer problem and then a mathematical model has been written for the drying process which also includes the convective term. Using the experimental data, the coefficient included in the mathematical model was found by the extremal method. So, an inverse problem was solved. The accuracy of the model was checked by comparing with the experimentally obtained temperature values after solving a direct heat transfer problem in the given conditions. Good correlation between the obtained model results and the experimental results shows the accuracy of the mathematical model.
- Open Access
- 1 September 2019
- Published by IOP Publishing in IOP Conference Series: Materials Science and Engineering
- Vol. 595 (1) , 012015
Research on the fire endurance of the silica sand membrane
In this study, the fire endurance, density, low-temperature deflection, penetration and ring-ball method tests of the membrane produced throughout the mixing of Bitumen, IPP (isotactic polypropylene), LDPE (Low-density polyethylene) calcite and silica sand at varying proportions were performed and the test results were compared with the membrane used today. In our study, TS 120 EN 1427, TS 118 EN 1426, TS EN 13501-1, KG-TL-029 testing methods were used. The density of the silica material we have obtained was measured at 1.3782 g/cm3, the endurance of the material was found out to be better, softening temperature was measured at 155°C, brittleness temperature at -7°C and the fire class was found out to be as E.
- 1 September 2019
- Published by IOP Publishing in IOP Conference Series: Materials Science and Engineering
- Vol. 595 (1) , 012015
Open AccessResearch on the fire endurance of the silica sand membrane
In this study, the fire endurance, density, low-temperature deflection, penetration and ring-ball method tests of the membrane produced throughout the mixing of Bitumen, IPP (isotactic polypropylene), LDPE (Low-density polyethylene) calcite and silica sand at varying proportions were performed and the test results were compared with the membrane used today. In our study, TS 120 EN 1427, TS 118 EN 1426, TS EN 13501-1, KG-TL-029 testing methods were used. The density of the silica material we have obtained was measured at 1.3782 g/cm3, the endurance of the material was found out to be better, softening temperature was measured at 155°C, brittleness temperature at -7°C and the fire class was found out to be as E.
- Open Access
- 1 September 2019
- Published by IOP Publishing in IOP Conference Series: Materials Science and Engineering
- Vol. 595 (1) , 012034
A review of new technologies in valve systems of internal combustion engines and their effects
Efficient use of energy is becoming increasingly crucial due to global warming and depleting fossil fuel sources. Automotive manufacturers aim for fuel economy, reduction of exhaust emissions, durability, volume and weight reduction, and power and torque increase in internal combustion engines. One of the most important developments in this field is the new technologies in classic valve systems. There are still ongoing studies in this field as the gas exchange processes affects all engine parameters. In this review; the principles of new technologies used on classic valve systems and their effects on engines have been investigated. The research has shown; new mechanism technologies used on valve systems have positive effects on engine performance, power, torque and fuel consumption.
- 1 September 2019
- Published by IOP Publishing in IOP Conference Series: Materials Science and Engineering
- Vol. 595 (1) , 012034
Open AccessA review of new technologies in valve systems of internal combustion engines and their effects
Efficient use of energy is becoming increasingly crucial due to global warming and depleting fossil fuel sources. Automotive manufacturers aim for fuel economy, reduction of exhaust emissions, durability, volume and weight reduction, and power and torque increase in internal combustion engines. One of the most important developments in this field is the new technologies in classic valve systems. There are still ongoing studies in this field as the gas exchange processes affects all engine parameters. In this review; the principles of new technologies used on classic valve systems and their effects on engines have been investigated. The research has shown; new mechanism technologies used on valve systems have positive effects on engine performance, power, torque and fuel consumption.
- Open Access
- 1 September 2019
- Published by IOP Publishing in IOP Conference Series: Materials Science and Engineering
- Vol. 595 (1) , 012056
Investigation of the effect of drying air temperature on drying time by numerical method for textile bobbins
Drying of the yarn bobbins in the textile sector is a very important stage among the textile finishing processes. Because this process takes a lot of time and causes a large amount of energy consumption. In this study, drying times for different drying air inlet temperatures at a constant drying air pressure were determined with the help of mathematical model for textile bobbin drying process. Finite difference method was used in mathematical model solution and the effect of different space and time steps were taken into consideration during the solution.
- 1 September 2019
- Published by IOP Publishing in IOP Conference Series: Materials Science and Engineering
- Vol. 595 (1) , 012056
Open AccessInvestigation of the effect of drying air temperature on drying time by numerical method for textile bobbins
Drying of the yarn bobbins in the textile sector is a very important stage among the textile finishing processes. Because this process takes a lot of time and causes a large amount of energy consumption. In this study, drying times for different drying air inlet temperatures at a constant drying air pressure were determined with the help of mathematical model for textile bobbin drying process. Finite difference method was used in mathematical model solution and the effect of different space and time steps were taken into consideration during the solution.