Cemal Keles

Sweep Frequency Response Analysis (SFRA) is one of the commonly used techniques for evaluating transformer winding integrity. However, the interpretation of the SFRA results remains qualitative and depends on expert evaluation. In this study, a quantitative, measurement-oriented diagnostic framework based on the advanced normalized Multi-Domain Fault Deviation Index (MFDI) is proposed. Within this method, magnitude, phase, impedance, and resonance-based descriptors are integrated in a unified formulation. The proposed method combines frequency band decomposition with multi-domain feature fusion to capture both local and global spectral deviations. This framework is experimentally validated under both intra-phase (U–N) and inter-phase (V–W) fault scenarios using SFRA measurements. The results demonstrate that the proposed MFDI captures the obvious difference between healthy and faulty states. For intra-phase fault scenarios, the proposed method provides quantitative differentiation among the investigated fault conditions despite the high similarity of their corresponding spectral signatures. In contrast, inter-phase faults generate more pronounced global spectral deviations, leading to highly separable feature representations for Support Vector Machine (SVM) and Random Forest (RF) classifiers. These findings demonstrate that the proposed approach advances SFRA from a predominantly qualitative interpretation technique toward a quantitative, interpretable, and machine-learning-compatible diagnostic methodology suitable for intelligent transformer condition monitoring.

Selection of suitable metaheuristics during the optimization process is a concern in real-time application, and the autonomous selection, known as hyper-heuristic selection, is requirement for intelligent system application. This study adopts a hyper-heuristic selection approach for the Neural Architecture Search (NAS) application and introduces a data-driven hyper-heuristic neuroevolutionary scheme. This scheme employs a maximum likelihood selection from a set of candidate metaheuristic according to their performance scores. The hyper-heuristic selector algorithm automatically and concurrently selects the most suitable metaheuristic algorithm while optimizing neural architecture of a deep neural network (DNN) model. As candidates for metaheuristic optimization algorithm, Atom Search Optimization Algorithm (ASO), Geometric Mean Optimizer (GMO), Multi-Verse Optimizer (MVO), Flow Direction Algorithm (FDA), and Sine Cosine Optimization (SCO) algorithms, inspired by different physical laws, were utilized. We demonstrate effectiveness of the performance of the proposed hyper-heuristic concurrent algorithm selector in optimal architecture DNN model design for the residential cooling load estimation problem. The proposed hyper-heuristic neuroevolutionary allows evolving to the best generalizing DNN model that well suits the modeling complexity requirements of the energy efficiency dataset. This experimental study reveals that the proposed hyper-heuristic neuroevolutionary can enhance performance of optimal DNN models compared to performances of solo metaheuristic neuroevolutionary algorithms.

Due to the complexity of climate systems, data-driven modeling based on observed time series data is essential for predicting future climatic trends. This study aims to improve the long-term global temperature anomaly forecast performance of Long Short-Term Memory (LSTM) based neural network models. Although several LSTM variants and hybrid architectures have been suggested for time series data prediction problems, the long-term forecast performance of these models may not be satisfactory in practice. To address solution of these problems, firstly, authors focused on evaluating the forecast performance of models and suggested performance and test assessment procedures. Secondly, authors suggest an Additive Twin LSTM (AT-LSTM) model that can improve the forecast performance for the global temperature anomaly. Our test on the Berkeley Global Temperature Anomaly dataset demonstrates that the proposed AT-LSTM can improve performance relative to conventional LSTM variants in long-term forecasting. Authors observed that global temperature trend projections of the AT-LSTM models for 20 years in future are consistent with expectations of climate organizations and projections in other works. The AT-LSTM models forecasted an average of 1.415 °C with ± 0.073 °C error in the year 2042 and this indicates the strong potential of major climate changes in the near future of Earth.

Bode’s ideal loop models have been utilized in control system designs to obtain performance robustness for the Direct Current (DC) gain variations. However, effects of crossover frequency and fractional order on control optimality of Bode's ideal loop reference models have not been sufficiently discussed, and it may raise a question of whether a control system design based on Bode’s ideal loop reference model is optimal. In this regard, this study revisits Bode’s ideal loops to investigate Integral Square Error (ISE) optimality of Bode’s ideal loops. For this purpose, the ISE optimality of Bode’s ideal loop is theoretically investigated by using Parseval’s theorem, and an ISE optimality condition is suggested in terms of crossover frequency and fractional order. Then, a generic controller family, which achieves an exact matching with the characteristics of the ISE optimal Bode’s ideal loop, was implemented for inverse fractional-order control of a class of minimum phase fractional order plant models. The authors present a numerical experiment on an ISE optimal Bode’s ideal loop inverse (OBILI) controller, and the ISE optimality and performance robustness of the designed control system were investigated for closed-loop control of a heat furnace system model via Monte Carlo simulations.

İklim modelleri, küresel ısınma ve iklim değişimi konularında elde edilen verilerin analizi ile iklim değişimlerinin yakın zamanlı gelişimi konusunda tahminlere olanak tanır. Bu çalışmada, küresel ısınma ve iklim değişimi konularında kullanılan enerji denge modelinin temelleri ve bu model yardımı ile yapılabilen analizler incelenmektedir. Bu basitleştirilmiş enerji denge modeli, güneşten dünyaya gelen ışıma enerjisinin dünyadan uzaya giden ışınım enerjisi ile dengelenmesi durumunu inceler. Bu bağlamda, küresel ısınmaya etki eden faktörlerin, denge durumu üzerinde etkisi olan birer radyatif (ışınımsal) zorlama olarak incelenmesine imkân sağlar. Böylece çok karmaşık ve dinamik olarak değişen meteorolojik ve jeolojik etmenlere bağlı olan iklim değişiminin, enerji bütçesi temelinde basitleştirilmiş analizlerini mümkün kılar. Bu çalışmada, Stefan-Boltzmann ve Wien Yer Değiştirme gibi temel ışınım kanunları çerçevesinde, güneşten gelen ışıma enerjisinin dünya tarafından emilim ve yansıma oranları dünya atmosferinden uzaya enerji çıkışını sağlayan ve elektromanyetik spektrumun uzun dalga boyunda yer alan atmosferik pencere kavramları incelenmiştir. Bu çerçevede, insan aktiviteleri ile değişebilme potansiyeli olan Albedo (yansıtabilirlik) radyatif zorlaması ve atmosferdeki CO2 molekülü birikimine bağlı radyatif zorlama etkileri açıklanmıştır. Sunulan temel kavramlar ve modellerin kolay anlaşılabilir olması için açıklayıcı örnekler ve verilere bağlı hesaplama ve analizler sunulmuştur. Bu makalenin küresel ısınma ve iklim değişimi konularında bazı temel kavramların açıklandığı eğitsel bir kaynak niteliği taşıması amaçlanmıştır.

The generalized classical method (GCM) was developed on the basis of the differential Taylor transform method, which was first used by the Ukrainian scientist G.E. Pukhov. In this study, some applications of the GCM to the analysis of transient events in simple electrical circuits are examined. It is shown that, if the solution can be decomposed into a steady-state part and a transient part, the use of the GCM becomes more effective, and the transient regime of the circuit solution can be analyzed without a full solving process of differential equations. The efficiency of the proposed approximation method is illustrated by comparing it with prior arts on similar problems. The results reveal that the proposed method is simple and can be applied successfully for the analysis of both linear and nonlinear problems in mathematical physics by similar solving procedures.

Multi-storey buildings with occupied highly technological instruments have more attraction possibilities of a lightning stroke than ordinary residential buildings. A protection system against lightning is necessary for these buildings and the level of the protection system performance is an issue to be examined. In this study, a protection system for a three-storey building is simulated in Alternative Transient Program-Electromagnetic Transients Program (ATP-EMTP). Distributed-parameter line models are used for the conductors of the protective system by modelling vertical conductors as nonuniform lines. The currents flowing in the critical conductors and the voltages at some critical points resulting by a lightning stroke are computed. The conventional lighting parameters required for structural protection such as lightning peak current, current rise time and time interval for the current flowing are determined. The effect of the radius of conductors in the protection system on the surge response is investigated.

Pamukkale University Journal of Engineering Sciences

In recent years, the energy system and the environment have faced major problems