A fuzzy software prototype for spatial phenomena: case study precipitation distribution

Download

index.pdf

Date

2010

Author

Yanar, Tahsin Alp

Metadata

Show full item record

Item Usage Stats

187
views

314
downloads

As the complexity of a spatial phenomenon increases, traditional modeling becomes impractical. Alternatively, data-driven modeling, which is based on the analysis of data characterizing the phenomena, can be used. In this thesis, the generation of understandable and reliable spatial models using observational data is addressed. An interpretability oriented data-driven fuzzy modeling approach is proposed. The methodology is based on construction of fuzzy models from data, tuning and fuzzy model simplification. Mamdani type fuzzy models with triangular membership functions are considered. Fuzzy models are constructed using fuzzy clustering algorithms and simulated annealing metaheuristic is adapted for the tuning step. To obtain compact and interpretable fuzzy models a simplification methodology is proposed. Simplification methodology reduced the number of fuzzy sets for each variable and simplified the rule base. Prototype software is developed and mean annual precipitation data of Turkey is examined as case study to assess the results of the approach in terms of both precision and interpretability. In the first step of the approach, in which fuzzy models are constructed from data, "Fuzzy Clustering and Data Analysis Toolbox", which is developed for use with MATLAB, is used. For the other steps, the optimization of obtained fuzzy models from data using adapted simulated annealing algorithm step and the generation of compact and interpretable fuzzy models by simplification algorithm step, developed prototype software is used. If the accuracy is the primary objective then the proposed approach can produce more accurate solutions for training data than geographically weighted regression method. The minimum training error value produced by the proposed approach is 74.82 mm while the error obtained by geographically weighted regression method is 106.78 mm. The minimum error value on test data is 202.93 mm. An understandable fuzzy model for annual precipitation is generated only with 12 membership functions and 8 fuzzy rules. Furthermore, more interpretable fuzzy models are obtained when Gath-Geva fuzzy clustering algorithms are used during fuzzy model construction.

Subject Keywords

Geographic Information Systems.

URI

http://etd.lib.metu.edu.tr/upload/12612596/index.pdf
https://hdl.handle.net/11511/20027

Collections

Graduate School of Natural and Applied Sciences, Thesis

Suggestions

OpenMETU
Core

A Modified Inverse Eigensensitivity Method for Large Finite Element Models Unlu, Dogus; Ciğeroğlu, Ender; Özgen, Gökhan Osman (2016-01-28) Finite element models should represent the dynamic behavior of real structures accurately to be subsequently used in design purposes. Therefore, finite element model updating methods have been developed in order to decrease the difference between analytical model and modal test results. In this paper, inverse eigensensitivity method as a sensitivity-based model updating method is summarized. Inverse eigensensitivity method with improved sensitivity computation which decreases the total calculation time of t...
Investigation of housing valuation models based on spatial and non-spatial techniques Boza, Ertuğrul; Düzgün, H. Şebnem; Türel, Ali; Department of Geodetic and Geographical Information Technologies (2015) The aim of this thesis is to develop hedonic housing valuation models based on spatial (SAR-simultaneous spatial autoregression and GWR - geographically weighted regression) and non-spatial (OLS - ordinary least squares) techniques, to compare the performances of these models and to investigate significant factors affecting housing value. The developed housing valuation models were tested at the Çankaya and Keçiören districts of Ankara province, Turkey. The results of the analyses revealed that significant ...
A novel modal superposition method with response dependent nonlinear modes for periodic vibration analysis of large MDOF nonlinear systems Ferhatoglu, Erhan; Ciğeroğlu, Ender; Özgüven, Hasan Nevzat (Elsevier BV, 2020-01-01) Design of complex mechanical structures requires to predict nonlinearities that affect the dynamic behavior considerably. However, finding the forced response of nonlinear structures is computationally expensive, especially for large ordered realistic finite element models. In this paper, a novel approach is proposed to reduce computational time significantly utilizing Response Dependent Nonlinear Mode (RDNM) concept in determining the steady state periodic response of nonlinear structures. The method is ap...
A Hybrid Computational Method Based on Convex Optimization for Outlier Problems: Application to Earthquake Ground Motion Prediction Yerlikaya-Ozkurt, Fatma; Askan Gündoğan, Ayşegül; Weber, Gerhard-Wilhelm (Vilnius University Press, 2016-01-01) Statistical modelling plays a central role for any prediction problem of interest. However, predictive models may give misleading results when the data contain outliers. In many real -world applications, it is important to identify and treat the outliers without direct elimination. To handle such issues, a hybrid computational method based on conic quadratic programming is introduced and employed on earthquake ground motion dataset. This method aims to minimize the impact of the outliers on regression estim...
A modified applied element model for the simulation of plain concrete behaviour Soysal, Berat Feyza; Arıcı, Yalın; Tuncay, Kağan (2022-08-01) A modified applied element model to simulate the behaviour of plain concrete continuum structures including discrete cracking is proposed in this study. In the classical applied element model, Poisson effects are fully ignored. To remediate this issue, diagonal elements are introduced to include the Poisson effect, and the constitutive parameters are rigorously determined using the Cauchy-Born rule and the hyper-elastic theory. The formulation is validated for linear elastic problems and the consistency and...

Citation Formats

T. A. Yanar, “ A fuzzy software prototype for spatial phenomena: case study precipitation distribution,” Ph.D. - Doctoral Program, Middle East Technical University, 2010.