Detection of Fm Transmitter Device Malfunctions With Artificial Neural Networks
No Thumbnail Available
Date
2024
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Bu çalışma, günümüzde popülerliği yüksek olan yapay sinir ağlarını kullanarak FM verici cihaz arızalarını tespit etme üzerinedir. Yapay sinir ağları, 1940'lı yıllarda ortaya çıkan, matematiksel formüller ile biyolojik sinir ağlarının modelleme işlemidir. Birçok alanda kullanılan bu yöntem karmaşık problemlerin çözümünde güvenli sonuçlar vermektedir. Avantajlarının yanında dezavantajları da bulunan Yapay sinir ağları (YSA) ile yapılan bu çalışmada elektronik bir cihaz olan FM vericinin parametreleri kullanılmıştır. Veri seti MATLAB programının içinde bulunan YSA modelleri ile denenmiş ve en iyi sonuç veren yöntem bulunmuştur. Elektronik cihaz arızalarını tespit etmede kullanılabilecek bir yöntem mi sorusunun cevabını aramaktayız. FM verici iki kısımdan oluşur. Bunlar; Exciter ve Amplifikatördür. Exciter olarak; İDİL elektronik firması tarafından üretimi yapılan AFM-30 isimli cihaz kullanılmıştır. Bu çalışma Exciter ünitesinde meydana gelen ve gelebilecek olan arızaların tespiti, Yapay sinir ağı modelleri kullanılarak yapılmıştır. Yapay sinir ağı mimarisinde, 12 adet girdi verisine karşılık binary sayı sistemi kullanılarak kodlanmış şekilde 3 adet çıktı verisi üretmesi sağlanmıştır. Bu tez çalışmasında amplifikatörün ne olduğundan bahsedilmiş ancak konu dışı olduğu için ayrıntılara girilmemiştir. İlkel yöntemlerin dışında teknolojinin gelişmesi de göz önüne alınarak farklı yöntem ve tekniklerin bulunmaya çalışıldığı günümüz dünyasında bir başka yöntem olan YSA ile de arızaların tespit edilebildiğini söyleyebiliriz.
This study is about detecting FM transmitter device malfunctions using artificial neural networks, which are highly popular today. Artificial neural networks are the modeling process of biological neural networks with mathematical formulas, which emerged in the 1940s. This method, used in many areas, provides safe results in solving complex problems. In this study conducted with Artificial Neural Networks (ANN), which has both advantages and disadvantages, the parameters of an FM transmitter, which is an electronic device, were used. The data set was tested with ANN models included in the MATLAB program and the method that gave the best results was found. We are looking for the answer to the question of whether it is a method that can be used to detect electronic device malfunctions. The FM transmitter consists of two parts. These; Exciter and Amplifier. As Exciter; A device named AFM-30, produced by İDİL electronic company, was used. In this study, the detection of malfunctions that occurred and may occur in the Exciter unit was carried out using Artificial neural network models. In the artificial neural network architecture, it is possible to produce 3 output data, encoded using the binary number system, in response to 12 input data. In this thesis, what an amplifier is is mentioned, but details are not given because it is off-topic. In today's world, where different methods and techniques are being tried to be found, considering the development of technology in addition to primitive methods, we can say that malfunctions can be detected with ANN, which is another method.
This study is about detecting FM transmitter device malfunctions using artificial neural networks, which are highly popular today. Artificial neural networks are the modeling process of biological neural networks with mathematical formulas, which emerged in the 1940s. This method, used in many areas, provides safe results in solving complex problems. In this study conducted with Artificial Neural Networks (ANN), which has both advantages and disadvantages, the parameters of an FM transmitter, which is an electronic device, were used. The data set was tested with ANN models included in the MATLAB program and the method that gave the best results was found. We are looking for the answer to the question of whether it is a method that can be used to detect electronic device malfunctions. The FM transmitter consists of two parts. These; Exciter and Amplifier. As Exciter; A device named AFM-30, produced by İDİL electronic company, was used. In this study, the detection of malfunctions that occurred and may occur in the Exciter unit was carried out using Artificial neural network models. In the artificial neural network architecture, it is possible to produce 3 output data, encoded using the binary number system, in response to 12 input data. In this thesis, what an amplifier is is mentioned, but details are not given because it is off-topic. In today's world, where different methods and techniques are being tried to be found, considering the development of technology in addition to primitive methods, we can say that malfunctions can be detected with ANN, which is another method.
Description
Keywords
Elektrik ve Elektronik Mühendisliği, Electrical and Electronics Engineering
Turkish CoHE Thesis Center URL
WoS Q
Scopus Q
Source
Volume
Issue
Start Page
End Page
107