Vol. 5 No. 3 (2026): December 31
Research Articles
-
Article: 55Published: 20.10.2026 pp. 176-193
HPPC Derived Feature Augmented DNN for SOC Estimation in Electric Vehicles
AbstractAccurate State of Charge (SOC) estimation is a critical function for ensuring the safety, performance, and longevity of lithium-ion batteries used in electric vehicles. Traditional model-based estimation methods—such as coulomb counting and Kalman filtering—often suffer from cumulative errors, parameter drift, and poor adaptability under varying temperature and dynamic load conditions. In contrast, data-driven approaches provide a promising alternative by learning complex nonlinear relationships directly from experimental data without relying on predefined equivalent circuit or electrochemical models. This study introduces an HPPC-derived feature-augmented deep neural network (DNN) framework that leverages the strengths of data-driven learning for accurate and robust SOC estimation. The proposed DNN is trained exclusively on Hybrid Pulse Power Characterization (HPPC) data and enhanced through feature engineering that captures voltage recovery dynamics, internal resistance, and temperature effects. Unlike conventional HPPC-based approaches, the model demonstrates strong generalization by accurately predicting SOC during EPA Urban Dynamometer Driving Schedule (UDDS) cycles—despite no exposure to driving-cycle data during training. Experimental validation on a 14-cell lithium-ion battery pack, managed by an L9963E-based battery management system, shows that the proposed model achieves a root mean square error (RMSE) reduction of 0.72% compared to traditional estimation techniques. These results confirm that data-driven DNN models trained on HPPC-derived features can offer high-accuracy, scalable, and generalizable SOC estimation solutions suitable for real-time deployment in next-generation automotive battery management system (BMS) platforms.
-
Article: 57Published: 20.10.2026 pp. 194-204
Impact of Oxygen Concentration on High-Load PEMFC Efficiency: A Simulink-Based Study
AbstractEnergy consumption and the utilization of sustainable energy sources are of great importance for the economic independence and environmental stability of nations. Turkey, as a country heavily dependent on energy imports, faces a critical need to develop domestic and renewable energy resources. The low carbon emissions and environmentally friendly nature of hydrogen have attracted significant attention within the scope of international agreements such as the Kyoto Protocol. Proton Exchange Membrane Fuel Cells (PEMFCs) stand out among clean energy technologies due to their low operating temperatures (60–80°C), high power density, and zero-emission potential. However, concentration losses occurring on the cathode side under high load conditions represent a significant problem that considerably reduces PEMFC efficiency and has not been sufficiently addressed in the literature.
This study investigates the effect of inlet air O₂ on PEMFC efficiency under different load conditions using a MATLAB-Simulink simulation model. The results demonstrate that increasing the oxygen concentration under 50% load conditions yields only a marginal efficiency improvement of approximately 3%. In contrast, under 90% load conditions, increasing the O₂ enrichment level from 21% to 100% improves efficiency by approximately 10%, and this gain can reach up to 15% with additional pressurization measures. This finding clearly demonstrates that increasing oxygen concentration provides a meaningful contribution only under high load conditions, while such intervention is unnecessary within the optimal load range. The results present a practical strategy for improving PEMFC efficiency in heavy-load applications such as intercontinental cargo ships and freight trains, and directly contribute to the literature on PEMFC systems operating under high load conditions. -
Article: 61Published: 20.10.2026 pp. 205-225
Optimal Sizing and Placement of Li-ion BESS for Virtual Inertia and Frequency Stability Enhancement in Grids with High Solar Penetration
AbstractThe global trend of large-scale high-capacity photovoltaic power generation has led to high instability due to inherent intermittency of solar energy and lack of mechanical inertia. Battery energy storage systems (BESS) as a solution can provide quick-response virtual inertia emulation (VIE). This study focused on the optimal combined sizing and location of lithium-ion BESS to achieve optimal virtual inertia supply and improve the frequency stability for grid networks with up to 40%-PV penetration. The optimization scheme aims to reduce the Life-Cycle Cost (LCC) and system penalties related to frequency deviations, the maximum frequency deviation and maximum allowable rate of change of frequency using the Teaching-Learning Based Optimization (TLBO). Simulations on modified IEEE 14 and 30 benchmark systems showed quantifiable benefits in optimal sizing of the assigned rating and capacities, namely the 44.4 MW and 193.9 MWh at Bus 8 on the 14-bus system, leading to significant attenuation of frequency deviation; in the 30-bus test case during a generator outage scenario, the frequency deviation was reduced from 0.503 Hz without BESS to 0.312Hz with BESS, and recovery time was reduced from 25s to 8s. In addition, the reliability indicators were improved, as shown with a decrease of the SAIDI index from 180 h/yr to 160 h/yr.
-
Article: 63Published: 20.10.2026 pp. 226-239
Prospects and Challenges of Deploying Solar-Wind Hybrid Energy Systems for Off-Grid Electrification in Zimbabwe
AbstractZimbabwe has a long history of electricity shortages especially in off-grid and rural regions where extension of grids is costly. Although the solar photovoltaic systems are quite common, their intermittency restricts regular power supply. The combination of wind and solar power can be more reliable due to the complementary nature of the resources. The paper evaluates these opportunities and challenges of implementing solar-wind hybrid energy systems to off-grid electrification in Zimbabwe, including the resource potential, the main strengths, and the obstacles, including the high cost, inaccessible wind data, and policy limitations.
-
Article: 65Published: 20.10.2026 pp. 240-251
Numerical Analysis and Simulation Approach of Copper Antimony Sulfide Based Thin Film Solar Cell with Dual Heterojunction
AbstractThis paper presents a SCAPS-1D numerical investigation of a Copper Antimony Sulfide (CuSbS2)-based thin-film solar cell using a dual-heterojunction n-ZnSe/p-CuSbS2/p+-BMO structure. The influence of ZnSe window-layer thickness and ferroelectric BSF materials including BFO, BTO and BMO was evaluated under AM1.5G illumination using literature-based material parameters. Under the adopted idealized simulation conditions, the optimized BMO-based structure produced Voc = 1.257 V, Jsc = 46.819 mA/cm2, FF = 83.21% and PCE = 48.96%. Because this result is unusually high for a CuSbS2-based thin-film device, it is interpreted as a numerical upper-limit case that requires further validation using complete optical, recombination and experimental device analysis.
-
Article: 64Published: 20.10.2026 pp. 252-282
End to end prediction of optimal power flow in sustainable power system using physics-informed spatiotemporal graph neural network
AbstractThis paper proposes a physics-informed spatiotemporal graph neural network (PIStGNN) for end-to-end ACOPF prediction in sustainable power systems. The proposed framework embeds physical power flow constraints, grid topology, and temporal dynamics of renewable generation. The model is validated on IEEE 33-, 57-, and 118-bus systems using time-series 60,048 datasets generated in Pandapower. All experiments are implemented in Python and performance are evaluated using prediction accuracy, physical feasibility, and computational efficiency. The results show that system voltage profiles remain within acceptable operational limits of 0.96–1.04 p.u. across all scenarios. The system efficiency analysis reveals that moderate renewable penetration reduces transmission losses, while excessive penetration increases losses due to reverse power flow and congestion in larger systems. The proposed PIStGNN achieves prediction errors on the order of 10⁻² p.u. for voltage magnitude and angle while delivering speedups of up to 10,135× compared with Newton–Raphson-based solvers. Furthermore, physics-informed recurrent graph models achieve minimum uncertainty values of 1.06 × 10⁻³, 2.88 × 10⁻³, and 1.19 × 10⁻³ for the IEEE 33-, 57-, and 118-bus systems, respectively. The results demonstrate that integrating physics-informed learning with spatiotemporal graph modeling provides an efficient surrogate for ACOPF prediction in sustainable power systems.
-
Article: 75Published: 20.10.2026 pp. 283-294
Air Temperature Forecasting Using a Long Short-Term Memory Model: A Case Study in Beykoz, Istanbul
AbstractHuman activities contribute to the production of greenhouse gases, which in turn lead to a rise in temperature and an increase in carbon footprint. With increasing frequency of extreme temperature events due to urban heat island effects and climate change, short-term high-resolution temperature forecasting is critical for localized energy management and urban climate adaptation. In this study, real-world atmospheric data, including as wind direction and speed, temperature, humidity, and precipitation amount are collected via the PCE-FWS 20N weather station in Beykoz, Istanbul. Applying machine learning models to actual atmospheric data demonstrates the potential for enhanced environmental predictions and promotes informed decision-making to ensure a sustainable future. A Long Short-Term Memory (LSTM) machine learning model with optimization methods ADAM (Adaptive Moment Estimation), NADAM (Nesterov-accelerated Adaptive Moment Estimation), and RMSprop (Root Mean Square Propagation) are applied independently for different epoch sizes. A comparative evaluation of the selected optimization algorithms was performed using RMSE and R² metrics to identify the reliable configuration for the analyzed dataset
-
Article: 94Published: 20.10.2026 pp. 295-309
Vision-Based Positioning and Control of a Robotic Manipulator for Tomato Harvesting
AbstractAutomated harvesting of tomatoes requires reliable integration of machine vision, distance measurement, and robotic-manipulator control to achieve accurate and repeatable product positioning. This study presents a vision-based positioning and control system for a robotic manipulator designed for automatic tomato harvesting. The proposed system integrates a webcam, Raspberry Pi 4 for image processing, a microcontroller for control, an HC-SR04 ultrasonic sensor for distance measurement, and servo motors for manipulator actuation. Image-based tomato coordinates are used to generate the required angular positions of the manipulator joints, while distance information is used to support the approach motion toward the detected product. A closed-loop control concept based on the comparison between commanded and actual angular positions is also considered. Experimental evaluation was performed using red and green tomatoes to assess detection, harvesting, and complete work-cycle times. The detection and harvesting times were 4.932 s for the red tomato and 5.276 s for the green tomato, while the complete cycle times were 9.676 and 10.586 s, respectively. For an image resolution of 320 × 180 pixels, an example object position of x = 15 pixels and y = 42 pixels corresponded to servo angles of 115° for the base joint and 100° for the first joint. The results demonstrate the feasibility of integrating visual perception, distance measurement, and manipulator motion control using a relatively simple hardware architecture. The proposed approach provides a basis for further development of accurate, repeatable, and energy-efficient robotic tomato harvesting systems.
