Browsing by Author "Ozkaya, Suat Gokhan"

Now showing 1 - 6 of 6

An Automated Earthquake Classification Model Based on a New Butterfly Pattern Using Seismic Signals
(Pergamon-Elsevier Science Ltd, 2024) Ozkaya, Suat Gokhan; Baygin, Mehmet; Barua, Prabal Datta; Tuncer, Turker; Dogan, Sengul; Chakraborty, Subrata; Acharya, U. Rajendra
Background: Seismic signals are useful for earthquake detection and classification. Therefore, various artificial intelligence (AI) models have been used with seismic signals to develop automated earthquake detection systems. Our primary goal is to present an accurate feature engineering model for earthquake detection and classification using seismic signals.Material and model: We have used a public dataset in this work containing three categories: (1) noise, (2) P waves, and (3) S waves. P and S waves are used to define earthquakes. We have presented two applied use cases using this dataset: (i) earthquake detection and (ii) wave classification. In this work, a new textural feature extractor has been presented by using a graph pattern similar to a butterfly. Thus, this feature extraction function is named Butterfly pattern (BFPat). We have created a new feature engineering architecture by deploying BFPat, statistics, and wavelet packet decomposition (WPD) functions. The recommended BFPat and statistics have been applied to the wavelet bands created by WPD and the raw seismic signals. Multilevel features have been extracted from both frequency and space domains. The used dataset contains signals with three channels. Using these three channels, seven signals have been created. Seven feature vectors have been created from 7 input signals used in this study. The most meaningful/informative features from the generated feature set are then selected using the iterative neighborhood component analysis feature selector method. Seven chosen feature vectors have been considered as inputs of the two shallow classifiers: k nearest neighbors (kNN) and support vector machine (SVM). A total of 14 (=7 x 2) results have been obtained in the classification phase. A majority voting process was applied in the last phase to choose the best results and improve the classification performance.Results: We have presented two use cases for our new BFPat method in this work to obtain superior results. Our model reached an accuracy of 99.58% in detecting the earthquake detection and 93.13% accuracy in 3-class classifications of waves.Conclusions: Our recommended model has achieved over 90% classification performance for both cases. Also, we have presented the most valuable channel and combinations in our work. Our developed system is ready to be tested with a bigger database.
Evaluation and Comparison of Ultimate Deformation Limits for Rc Columns
(Elsevier Science Ltd, 2017) Ozdemir, Muhammed Alperen; Kazaz, Ilker; Ozkaya, Suat Gokhan
The ultimate deformation capacity assigned to structural elements of reinforced concrete structures is an essential parameter in the determination of their structural performance especially under seismic attack. Various ultimate deformation capacity limits were proposed for RC columns in the previous studies and were accommodated in the current building codes. However, the reliability of the existing deformation limits is still a matter of considerable debate. This study mainly focuses on the evaluation of the accuracy of the existing damage limits and aims to develop a new definition with a higher reliability in comparison to the existing limits. In this purpose, the study was composed of four major steps. Firstly, the existing building codes, standards, regulations and previous studies were reviewed and evaluated in terms of ultimate deformation criteria for RC columns. Secondly, actually tested sixty-nine RC columns were selected from PEER Structural Performance Database and were numerically modelled by using finite element method. The selected RC columns have different dimensions, aspect ratios, concrete and steel strength, longitudinal and transverse reinforcement amounts and axial load ratio. Third stage includes the comparison of the results obtained from the experimentally verified numerical models with the existing ultimate deformation limits to reveal the shortcomings of existing criteria. The numerical modelling facilitated the consistent comparison of strains and curvatures that are rarely available and difficult to measure in tests. A new concrete compressive strain prediction equation was proposed to determine the ultimate deformation capacity of rectangular RC columns. It was concluded that there is need for further comprehensive analytical and experimental studies on deformation limits of reinforced concrete columns.
Most Complicated Lock Pattern-Based Seismological Signal Framework for Automated Earthquake Detection
(Elsevier, 2023) Ozkaya, Suat Gokhan; Baygin, Nursena; Barua, Prabal D.; Singh, Arvind R.; Bajaj, Mohit; Baygin, Mehmet; Acharya, U. Rajendra
Background: Seismic signals record earthquakes and also noise from different sources. The influence of noise makes it difficult to interpret seismograph signals correctly. This study aims to develop a computationally lightweight, accurate, and explainable machine learning model for the automated detection of seismogram signals that could serve as an effective warning system for earthquake prediction.Material and method: We developed a handcrafted model for earthquake detection using a balanced dataset of 5001 earthquakes and 5001 non-earthquake signal samples. The model included multilevel feature extraction, selector-based feature selection, classification, and post-processing. Input signals were decomposed using tunable Q wave transform and fed to a statistical and textural feature extractor based on the most complicated lock pattern (MCLP). Four feature selectors were used to choose the most valuable features for the support vector machine classifier. Additionally, voted vectors were generated using iterative hard majority voting. Finally, the best model was chosen using a greedy algorithm.Results: The presented self-organized MCLP-based feature engineering model yielded 96.82% classification ac-curacy with 10-fold cross-validation using the seismic signal dataset.Conclusions: Our model attained high seismological signal detection performance comparable with more computationally expensive deep learning models. Our handcrafted explainable feature engineering model is computationally less expensive and can be easily implemented. Furthermore, we have introduced a competitive feature engineering model to the deep learning models for the seismic signal classification model.
A New Approach Based on Image Processing for Measuring Compressive Strength of Structures
(2017) Kazaz, İlker; Baygın, Mehmet; Özdemir, Muhammed Alperen; Ozkaya, Suat Gokhan
The compressive strength factor in civil engineering is a very important parameter used to determine the performance ofstructures. The stability of structures can be tested with this parameter which is used to measure the performance of concrete under differentloads. This parameter, which should be determined for the safety of the structures, is usually based on experimental analyses performed inthe laboratory environment. In this study, a new approach to compressive strength measurement in civil engineering is proposed. With thisapproach, which is based on image processing, measurement of compressive strength parameter of concrete samples taken from structuresis performed. For this purpose, images of concrete specimens with different strengths are taken and these images are divided into twogroups as training and test set. Then, image processing algorithms are applied to these images and the compressive strength of concretespecimens is calculated. It has been determined that the approach suggested in the test runs performed with an error rate of about 1-2%.
Yapay Sinir Ağı ve Görüntü İşleme Tabanlı Basınç Dayanımı Tahmini
(2021) Kazaz, İlker; Durur, Hülya; Ozkaya, Suat Gokhan; Bayğın, Mehmet
Ülkemizde çok çeşitli medeniyetlerden günümüze ulaşmış birçok eser mevcuttur. Tarihi yığma yapılar da, kültürel miras olarak kabul edilen bu yapıların arasında yer almaktadır. Bu nedenle bu tarihi yığma yapıların detaylı bir şekilde incelenmesi, bu yapılara ait numunelerin test edilmesi ve bu testlerden elde edilen bu bilgilerin dökümante edilmesi, bu yapıların gelecek nesillere sağlam bir şekilde aktarılabilmesi açısından oldukça önemli bir konudur. Gerçekleştirilen bu çalışmada, yığma yapılarda kullanılan taşların bilgisayarlı görme teknolojisi ile incelenmesi yapılmıştır. Bu amaçla taş ocaklarından alınan farklı niteliklere sahip taşların öncelikli olarak bir kamera aracılığıyla görüntüleri alınmıştır. Daha sonrasında her bir numunenin görüntüsü bilgisayar ortamına aktarılmış ve bu numunelere ait özellikler görüntü işleme yoluyla elde edilmiştir. Daha sonrasında bu numuneler laboratuvar ortamında test edilmiş ve dayanımları ölçülmüştür. Sonuç olarak laboratuvar ortamından elde edilen veriler ve görüntü işleme ile elde edilen sonuçlar karşılaştırılmış, önerilen bu görüntü işleme tabanlı analiz yöntemi kalibre edilmiştir. Önerilen yöntem traverten, andezit ve kireç taşı için sırasıyla %99.47, %93.89 ve %94.19 olarak belirlenmiştir. Yapılan bu çalışmalar neticesinde laboratuvar ortamında yapılan deneysel uygulamaların bilgisayar ortamında gerçekleştirilebilmesine olanak sağlanmıştır. Yine bu çalışmalar neticesinde oldukça zorlu ve uzun süren çalışmalar önemli bir ölçüde kısaltılmış, temel bir kamera aracılığıyla görüntüsü alınan numunelerin belirli özellikleri bilgisayar ortamında çok hızlı bir şekilde hesaplanabilmiştir.
Yığma Kemer Köprülerde Kullanılan Malzeme Özelliklerinin Tahribatsız Yöntemlerle Belirlenmesi
(2020) Kazaz, İlker; Kocaman, İrfan; Ozkaya, Suat Gokhan
Yığma yapıların hesabında yapıyı oluşturan harç, taş vb. malzemelerin özelliklerini içinde eriten tekbir homojenize malzeme kullanımı yaygın bir uygulamadır. Bu malzemenin basınç dayanımı ve elastikmodülü gibi özellikleri onu oluşturan birimlerin karakteristik değerlerinden hesaplanabilmektedir.Taş malzemesinin mekanik özelliklerini hasarsız deneyler yardımıyla belirlemek için Schmidt sertlikçekici ve ultrasonik ölçüm cihazları ile P dalga hızı oldukça yaygın kullanılmıştır. Bu çalışmadaErzurum ilinde bulunan tarihi Kireçli ve Kız köprüleri ele alınmıştır. Her iki köprüde yerindeçalışmalar gerçekleştirilerek köprüleri oluşturan taş malzemenin cinsi belirlenerek taş ocaklarındannumuneler alınmıştır. Kübik taş numuneleri fiziksel, indeks ve mekanik testlere tabi tutularak buözellikler arasında ilişki bağıntıları kurulmuştur. Taş numuneler üzerindeki çalışmalardan sonra üçtaş bloktan oluşan yığma birimleri üzerinde çalışmalar gerçekleştirilmiştir. Yığma birimlerin basınçdayanımları ile P dalga hızı arasında oldukça yüksek korelasyona sahip ilişki belirlenmiştir.