Loading...

Modeling of a Vague Problem (Floating Breakwater by BEM Analysed) Using ANFIS (Adaptive Neuro-Fuzzy Inference System) and Validation by Fuzzy Boundary Element Method

Fereydoon, Maryam | 2010

647 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 40568 (08)
  4. University: Sharif University of Technology
  5. Department: Mechanical Engineering
  6. Advisor(s): Abbaspour Tehrani, Majid
  7. Abstract:
  8. The thesis title is “Modeling of a vague problem (Floating Breakwater by BEM Analysed) using ANFIS (Adaptive Neuro-Fuzzy Inference System) and validation by Fuzzy Boundary Element Method”. In this set of material, more important definitions are described below. When some input data is linguistics, or we have lack of Information or Vagueness, Fuzzy toolbox can alleviate the problem. Fuzzy concept includes several fields and applications like; fuzzy logic, fuzzy sets, fuzzy control, fuzzy arithmetic and fuzzy inference, and the last two applications are used in this thesis. The Neural network is a black box, Neurons that can justify their actions and these artificial brains can grow. The representative example is a floating breakwater with undeterminate geometry. The BEM code renders FBW by solving Laplacian equation for Potential, and the output is a crisp set that should be fuzzified and modeled with ANFIS, Overall this set of numbers should be validated with analytic FBEM results. Neuro-Fuzzy Concept includes Two Approaches, first Neuro-Fuzzy models, and Neuro-Fuzzy Inference, that my research field is the next one. The learning Process of this provides many networks such as, ARIC, GARIC, ANFIS, NNDFR, FUN, NeuFuz, DAX, etc. I chose the ANFIS for these determinant set of data, generated by the mentioned example. Gauss-Green theorem is the basis of Boundary Element Method, in this numerical method, after fundamental Solution of governing equations, numerical solution can be generated. In this method three major element types are defined, Continuous linear, Discontinuous linear (Corner Problem) and Higher Order elements, in this thesis, the second type is used to analyse the hydrodynamical problem. The mentioned problem is a vague one, a vague problem can be discussed when, vagueness is in the boundary condition, geometry or constant parameters, etc. a sample problem in geometry is “Corner problem” (different boundary condition for each element). Why inverse problems are interesting and practically important is inverse problems are the problems that consist of finding an unknown property of an object, or a medium, from the observation of a response of this object, or medium, to a probing signal. Such problems are important in identification of flying objects (airplanes missiles, etc.), objects immersed in water (submarines, paces of fish, etc.), and in many other situations.
  9. Keywords:
  10. Neuro Fuzzy Network ; Boundary Element Method ; Adaptive Neuro-Fuzzy Inference System (ANFIS) ; Vague Problem ; Fuzzy Boundary Element Method (FBEM)

 Digital Object List

 Bookmark

...see more