Geri Dön

Uçuş mürettabatı için arıza teşhis uzman sistemi

Fault diagnosis expert system for flight crew

  1. Tez No: 127021
  2. Yazar: SEDEF ÖZÇELİK
  3. Danışmanlar: PROF. DR. ALPASLAN FIĞLALI
  4. Tez Türü: Yüksek Lisans
  5. Konular: Savunma ve Savunma Teknolojileri, Defense and Defense Technologies
  6. Anahtar Kelimeler: Belirtilmemiş.
  7. Yıl: 2002
  8. Dil: Türkçe
  9. Üniversite: İstanbul Teknik Üniversitesi
  10. Enstitü: Fen Bilimleri Enstitüsü
  11. Ana Bilim Dalı: Uzay Mühendisliği Ana Bilim Dalı
  12. Bilim Dalı: Belirtilmemiş.
  13. Sayfa Sayısı: 86

Özet

UÇUŞ MÜRETTEBATI İÇİN ARIZA TEŞHİS UZMAN SİSTEMİ ÖZET Bu tezde güdülen birinci amaç tamamen bilgisayara dayalı uçuş mürettebatı için bilgi bir arıza teşhis ve giderme US'i geliştirmektir. US'ler bilgiyi bir insanın yaptığı gibi işlemeye çalışan YZ programlan olarak, zamanın bilgiye olan açlığına karşı kullanılacak değerli araçlardır. Arıza giderme ve teşhis US'leri ise kendilerine duyulan ihtiyacın hiçbir zaman sona ermeyeceği US'ler olarak bütün endüstri alanlarında yaygın olarak kullanılmaya başlanmıştır. Tezde US kavramı tanımlanmış, yapısı, sınırlan, yararlan, uygulama alanları ve bilgi gösterim metotları konularına yer verilmiştir. Tezin amacı, problemin tanım, problemin alanı, anormal durumlarda uçuş mürettebatının görev yönetimi, uçuş mürettebatının arıza teşhis el kitabının yapısı, bilgi tabanlı pilot kabini yardımcı sistemi incelenmiştir. Tezin ikinci aşamasında geliştirilen arıza giderme US programı kullanılarak gerçek hayattan bir uygulama ortaya çıkarmak ve böylelikle bu geliştirilen programın bir çok ticari uçakta uçuş mürettebatı için kullanılabileceğini kanıtlamaktadır. Böylece pilot iş yükünün azaltılması ve operasyon etkinliğinin artırılması hedeflenmiştir. Günümüzde yeni nesil uçak tasarımı ve mevcut uçak modifikasyonlarında bu tür çözümlere gidilmektedir. US kurma araçlarından yazılım dili olarak, önerilen sistemin bilgi tabanlı olması gerektiğinden bir kural dili olan Prolog seçilmiştir. Bu amaca ALS Prolog US aracı kullanılarak hazırlanan bir programla ulaşılmıştır.

Özet (Çeviri)

FAULT DIAGNOSIS EXPERT SYSTEM FOR FLIGHT CREW SUMMARY Expert systems (ES) are computerised program tools that mimic the judgement of human experts in a well defined domain. ES are one area of artificial intelligence that is being developed and used many diagnostic tasks. The fault diagnosis of machine is a complicated engineering problem. It is highly related to the manufacture surrounding. The diagnosis ability of an ES is decided by the actual experiences and knowledges of its designer. For the diversity of machine and the difference of the manufacture surrounding, it is impossible for the designers to know all the facts of the machine that ES will be used to. So we can't expect that an ES has the diagnosis ability of a real expert at birth. Expert systems typically have a number of distinguishing components. The knowledge base is the component that contains the knowledge obtained from the domain expert. The most common way of representing knowledge is using rules. The inference engine is the component that manipulates the knowledge found in the knowledge base as needed to arrive at a result or solution. The user interface is the component that allows the user to query the system and receive the results of those queries. Many ES also have an explanation facility which explains why a question was asked or how a result or solution was obtained. There are three different mains of building an ES knowledge base, namely rubs, frames and semantic networks. Frames method is used mostly when the variables in the knowledge base have hierarchical structure. Since frames allow the structured representation of descriptive knowledge, they also facilitate analogical reasoning. Frame systems provide no direct methods for describing how the knowledge stored in frames is to be used. Semantic networks method is preferred when the variables of the knowledge base have a lot of different interrelations such as“is, can, has, etc.”. Semantic networks (also called associative networks), are another major form of knowledge representation. In such systems, knowledge is represented in the form of nodes and arcs, where nodes represent objects, classes, and events while arcs represent relationships that hold between the concept nodes. Production rules is preferred when the rules of the knowledge base can be expressed in the form of“IF,..., THEN”. Most of the developed ES uses production rules. In logic based systems, knowledge is represented as simple and complex predicate vistatements. Logic based systems rely on resolution theorem proving for their inference mechanisms. In these systems prescriptive knowledge is represented in the form of condition-action rules or procedures. Classifiers and classifier systems based on genetic algorithms are inductive learning systems. In these systems knowledge is represented as vectors or matrices of values to a set of parameters or a set of message lists. Genetic algorithms represent knowledge as vectors of values to a set of parameters or a set of message lists. GA based classifiers These systems support, besides inductive learning, conflict resolution, classification and learning control rules. Neural nets are composed of a large number of interconnected processing units. In these systems knowledge is represented by the totality of the connections. Neural networks support inductive learning, classification and conflict resolution. Prolog is a unique programming language. While other languages like BASIC, FORTRAN and PASCAL are classified as procedural languages, Prolog is a declarative language. In other languages the programmer must write the procedures of how the problem is to be solved in order to get a program working, whereas in Prolog computing procedures are built-in in the language. All the programmer has to do is give the necessary information about the problem domain and let Prolog find the answers to the questions asked by the programmer. Prolog is based on First Order Predicate Logic and the name Prolog derives from PROgramming in LOGic. First Order Predicate Logic is a well developed part of Symbolic Logic. A Prolog programmer is not a programmer but a logician, or in fact any one who can represent information in a logical way. Advanced in electronics and computer technology have a profound effect on modern aircraft. On flight deck electronic displays and computer controlled avionic devices are common features. Powerful computer technology is involved in the signal processing to generate display formats and to transmit signals to and from control devices and the remaining avionic components. However, the crew station as we know it today is under-going an even more profound chance, which will allow to make use of abstract human like knowledge and reasoning. Non-normal situations require the flight crew to devote more time and attention to the function related to the cause of the non-normal condition (for example, systems management if an engine has failed, or flight management ifa storm is encountered). However, it is rare that this function is the only one affected. Tasm mamagement has a large role to play in how the crew's attention is dispersed among the different functions. Checklists are used by the flight crew to properly configure an aircraft for safe flight and to ensure a high level of safety throughout the duration of the flight. In addition, the checklist provides a sequential framework to meet cockpit operational requirements, and it fosters cross-checking of the flight deck configuration among crew members. vuThis thesis examined expert system instead of checklists for non-normal procedures for a typical transport aircraft. For the purposes of this thesis, a typical transport aircraft (Boeing 737-400/500) checklists is used for application which is programmed in Prolog. vni

Benzer Tezler

  1. Tez No

    424235

    An evaluation of the benefits of simulator trainings for pilots by using physiological measurements

    Pilotlar için simülatör eğitiminin faydasının fizyolojik ölçümler kullanılarak değerlendirilmesi

    İSMAİL KUMPAS

    Yüksek Lisans

    İngilizce

    İngilizce

    2016

    Elektrik ve Elektronik MühendisliğiTürk Hava Kurumu Üniversitesi

    Elektrik-Elektronik Mühendisliği Ana Bilim Dalı

    YRD. DOÇ. DR. MEMİK TAYLAN DAŞ

  2. Tez No

    881060

    Airbus A320 side-stick kontrolcü sınırları, incelenmesi, diğer sistemlere uyarlanması

    Overall examinations on restrictions of the Airbus A320 side-stick controller and entegration of other systems

    HÜSEYİN GÜLHAN

    Yüksek Lisans

    Türkçe

    Türkçe

    2024

    Sivil HavacılıkErciyes Üniversitesi

    Havacılık Elektrik ve Elektroniği Ana Bilim Dalı

    DOÇ. DR. MEHMET KONAR

  3. Tez No

    909220

    Helikopter pilotları için titreşim azaltıcı koltuk izolatörü geliştirilmesi

    Developing a vibration reducing seat isolator for helicopter pilots

    EMRE CAN GÜLÜNAY

    Yüksek Lisans

    Türkçe

    Türkçe

    2024

    Havacılık ve Uzay MühendisliğiErciyes Üniversitesi

    Uçak Gövde Motor Bakım Ana Bilim Dalı

    DOÇ. DR. MUSTAFA SOYLAK

  4. Tez No

    791504

    Havacılıkta meteoroloji verilerinin uçuş emniyeti açısından ekip kaynak yönetimine etkileri

    The effects of meteorology data on crew resource management in terms of flight safety in aviation

    ALTUĞ AYKAN BAYAZITOĞLU

    Yüksek Lisans

    Türkçe

    Türkçe

    2023

    Meteorolojiİstanbul Gelişim Üniversitesi

    Havacılık Yönetimi Ana Bilim Dalı

    DR. ÖĞR. ÜYESİ HABİBE GÜNGÖR

  5. Tez No

    203097

    Helikopter uçucularında işitme kayıplarının incelenmesi

    Analysis of hearing loss among helicopter flight personnel

    KORAY ERGİN

    Yüksek Lisans

    Türkçe

    Türkçe

    2007

    Adli TıpAnkara Üniversitesi

    Disiplinlerarası Adli Tıp Ana Bilim Dalı

    PROF. DR. İZZET DUYAR

Geri Dön