Transient event classification using pmu data with deep learning techniques and synthetically supported training-set

Özgül SALOR-DURNA

This paper presents a research work which focuses on generating synthetic data to enrich the training-set of a deep learning (DL) based classification system to classify power system transient events using PMU frequency measurements. The synthetically improved training-set is shown to increase the classification performance compared to the case when only the actual-data training-set is used. The proposed classification system helps to reveal high-frequency transient variation information out of PMU measurements collected at a relatively much lower rate, especially when a small set of training- ...Daha fazlası

Makale 2023 Mühendislik Fakültesi IET Generation, Transmission & Distribution17 ( 6 ) 64 Görüntülenme

Curved beam generation and its experimental realization by rectangular prism with asymmetric polynomial back surface

Mirbek TURDUEV

With the discovery of self-accelerating beams, possibility of obtaining curved light beams in free space has been realized. These special beams paved the way for many new applications as well as the exploration of novel beam types. Recently, great research effort has been conducted to realize different types of curved beams such as photonic hook and airy beam. These curved types of beams are obtained by introducing structural asymmetry or applying non-uniform dielectric distribution to the input and output face of the structure. With this regard, we propose specially designed asymmetric struct ...Daha fazlası

Makale 2023 Mühendislik Fakültesi Physica Scripta98 ( 1 ) 79 Görüntülenme

Broadband, Stable, and Noniterative Dielectric Constant Measurement of Low-Loss Dielectric Slabs Using a Frequency-Domain Free-Space Method

Mehmet ERTUĞRUL

A broadband, stable, and noniterative free-space method is proposed for dielectric constant ε′r determination of low-loss dielectric slabs from reflection-only measurements through simple calibration standards (reflect and air). It is applicable for dispersive samples and does not require thickness information. Simulations of nondisperive and dispersive samples are performed to validate our method. Dielectric constant measurements of polyethylene and polyoxymethylene samples (9–11 GHz) are carried out to examine the accuracy of our method.

Makale 2022 Mühendislik Fakültesi IEEE Transactions on Antennas and Propagation70 ( 12 ) 65 Görüntülenme

Close link between Fabry–Pérot resonance and natural-resonance frequencies

Mehmet ERTUĞRUL

Mathematical and numerical analyses have been performed to examine the close link between Fabry–Pérot resonance and natural-resonance frequencies. For the mathematical analysis, the conditions resulting in minimum magnitudes of reflection coefficients in frequency-domain are derived for air-backed and metal-backed low-loss non-dispersive (or weakly dispersive) dielectric samples with relative complex permittivity εr and thickness L for free-space wave propagation at normal incidence. The close relation between Fabry–Pérot resonance and natural-resonance frequencies is demonstrated for three di ...Daha fazlası

Makale 2023 Mühendislik Fakültesi Waves in Random and Complex Media ( Published online: 23 Jan 2023 ) 48 Görüntülenme

Waveform Correlation Based Harmonic Voltage Contribution Determination of Iron and Steel Plants Supplied From PCC

Özgül SALOR-DURNA

In this research work, a new method which determines the individual harmonic voltage contributions of the EAF plants supplied from a point of common coupling (PCC) to the PCC voltage is presented. EAFs are one of the most significant sources of the harmonics, especially the uncharacteristic ones, therefore it is important to be able to discriminate the amount of individual contributions from the feeders of a PCC supplying multiple EAFs. The proposed method uses the relationship derived between the correlation coefficient of the PCC voltage and the feeder current waveforms and the harmonic volt ...Daha fazlası

Konferans Bildirisi 2022 Mühendislik Fakültesi IEEE Industry Applications Society Annual Meeting 31 Görüntülenme

Artificial Neural Network Model for Evaluating Parameters of Reflection-Asymmetric Samples From Reference-Plane-Invariant Measurements

Mehmet ERTUĞRUL

A technique based on artificial neural network (ANN) is proposed to extract the electromagnetic properties of reflection-asymmetric samples from reference-plane-invariant (RPI) scattering parameter measurements. It first determines reference plane transformation distances and then extracts the material properties. The number of neurons in the hidden layer of the ANN model was evaluated subject to accuracy and time constraints. We examined the conformity of the dataset of the ANN model and the required time for the training process by considering different numbers of neurons in the selected hid ...Daha fazlası

Makale 2023 Mühendislik Fakültesi IEEE Transactions on Instrumentation and Measurement72 62 Görüntülenme

Excitation wavelength dependent triple-mode photoluminescence of copper-based halides for advanced anti-counterfeiting

Mehmet ERTUĞRUL

New fluorescent materials with a low cost, hypotoxicity, and concealment are desired for the application of anti-counterfeiting. Herein, we report a CsCu2I3@Cs3Cu2I5 composite with a triple-mode photoluminescence (PL) feature by simply adjusting the excitation wavelengths, which are ascribed to the multiple excited states of different phases in the CsCu2I3@Cs3Cu2I5 composite. The broadband emission and high quantum yield (∼51) of the composite originate from the structure-oriented self-trapped excitons effect of Cs3Cu2I5 and CsCu2I3 phases. Moreover, the incorporation of polyethylene oxide (PE ...Daha fazlası

Makale 2023 Mühendislik Fakültesi APL Materials11 ( 7 ) 38 Görüntülenme

Temporal Reflection Response of Plane Waves Within an N-Layer Structure Involving Bianisotropy

Mehmet ERTUĞRUL

The temporal reflection response of an N -layer composite structure involving bianisotropic behavior is examined for the first time in the literature. Toward this goal, the wave matrix (WM) method, which is fundamental, relatively easier to apply, and compact and iterative suitable for analysis of multilayer structures, is first extended to the analysis of an interface separating two different media involving bianisotropy. Next, the reflection response of the whole structure is derived in terms of series form to better analyze its temporal behavior. Then, frequency-domain and time-domain respo ...Daha fazlası

Makale 2023 Mühendislik Fakültesi IEEE Transactions on Antennas and Propagation71 ( 7 ) 46 Görüntülenme

Waveform Correlation Based Harmonic Voltage Contribution Determination of Iron and Steel Plants Supplied From PCC

Özgül SALOR-DURNA

This paper presents a new waveform correlation based method which determines the individual harmonic voltage contributions of Electric Arc Furnace (EAF) plants supplied from a point of common coupling (PCC). The method is based on the waveform correlation computations between the PCC voltage and the feeder current at the individual harmonic frequency. PCCs supplying multiple EAF plants usually suffer from high voltage harmonic components due to their operation principles. A relationship between the correlation coefficient of the PCC voltage and the feeder current waveforms, and the harmonic vo ...Daha fazlası

Makale 2023 Mühendislik Fakültesi IEEE Transactions on Industry Applications59 ( 4 ) 24 Görüntülenme

Extraction Method for Unique Determination of Complex Permittivity from Graphical Analysis of Natural Frequencies

Mehmet ERTUĞRUL

A time-domain free-space extraction procedure is proposed to uniquely determine the relative complex permittivity ϵr (ω) based on two sets of natural frequencies of the metal-and air-backed dielectric samples. A graphical analysis is applied for evaluating whether unique ϵr (ω) can be extracted using such natural frequencies. The method is numerically validated using two dielectric samples with different ϵ (ω) for the ideal case (no noise) and under the influence of noise with different signal-to-noise ratios (SNRs). Keyword: graphical method; natural modes; parameter extraction; unique soluti ...Daha fazlası

Makale 2023 Mühendislik Fakültesi IEEE Transactions on Antennas and Propagation71 ( 9 ) 19 Görüntülenme

Classification of high frequency transient events buried in low frequency sampled PMU data

Özgül SALOR-DURNA

This paper presents a deep-learning based method for the classification of transient events of the power system using the data collected by a standard Phasor Measurement Unit (PMU). Although transient power system events are high-frequency variations of the voltages and currents, it leaves a trace in the PMU time-series data, which are collected at relatively much lower-frequencies. The main motivation of this study is to use low-frequency data recorded by PMU devices, widely used in the industry, for the classification of transient events, which would normally require the use of high-cost pow ...Daha fazlası

Konferans Bildirisi 2023 Mühendislik Fakültesi Proceedings 9 Görüntülenme

Machine learning enabled the design of compact and efficient wavelength demultiplexing photonic devices

Mirbek TURDUEV

In this paper, we introduce the design approach of integrated photonic devices by employing reinforcement learning known as attractor selection (AttSel). Here, we combined 3D FDTD with AttSel algorithm, which is based on artificial neural networks, to achieve ultra-compact and highly efficient wavelength demultiplexers with low crosstalk such as. The presented devices consist of SOI materials, which are compatible with complementary MOS technology. Consequently, the reinforcement learning is successfully applied to design smaller and superior integrated photonic devices. Keyword: inverse desig ...Daha fazlası

Konferans Bildirisi 2023 Mühendislik Fakültesi 2023 IEEE Photonics Conference (IPC) 18 Görüntülenme

Detaylı Arama

Filtreler

Koleksiyon [1]

Araştırmacılar [3]

BM Sürdürülebilir Kalkınma [3]

Fakültesi / Enstitütü [1]

Yayın Türü [2]

Yayımlanma Yılı [5]

Dil [1]

Kaynak [23]

Veri Tabanları [12]

İndex Türü [4]

Yayıncı [10]

Bölümü [1]

Ulusal/Uluslararası [1]

Konu Başlıkları [35]

Transient event classification using pmu data with deep learning techniques and synthetically supported training-set

Özgül SALOR-DURNA

Curved beam generation and its experimental realization by rectangular prism with asymmetric polynomial back surface

Mirbek TURDUEV

Broadband, Stable, and Noniterative Dielectric Constant Measurement of Low-Loss Dielectric Slabs Using a Frequency-Domain Free-Space Method

Mehmet ERTUĞRUL

Close link between Fabry–Pérot resonance and natural-resonance frequencies

Mehmet ERTUĞRUL

Waveform Correlation Based Harmonic Voltage Contribution Determination of Iron and Steel Plants Supplied From PCC

Özgül SALOR-DURNA

Artificial Neural Network Model for Evaluating Parameters of Reflection-Asymmetric Samples From Reference-Plane-Invariant Measurements

Mehmet ERTUĞRUL

Excitation wavelength dependent triple-mode photoluminescence of copper-based halides for advanced anti-counterfeiting

Mehmet ERTUĞRUL

Temporal Reflection Response of Plane Waves Within an N-Layer Structure Involving Bianisotropy

Mehmet ERTUĞRUL

Waveform Correlation Based Harmonic Voltage Contribution Determination of Iron and Steel Plants Supplied From PCC

Özgül SALOR-DURNA

Extraction Method for Unique Determination of Complex Permittivity from Graphical Analysis of Natural Frequencies

Mehmet ERTUĞRUL

Classification of high frequency transient events buried in low frequency sampled PMU data

Özgül SALOR-DURNA

Machine learning enabled the design of compact and efficient wavelength demultiplexing photonic devices

Mirbek TURDUEV