Forecasting of entropy production due to buoyant convection using support vector machines (SVM) in a partially cooled square cross-sectional room

Authors:
Yasin Varol;Hakan F. Oztop;Ahmet Koca;Engin Avci
Affiliations:
Department of Mechanical Education, Firat University, 23119 Elazig, Turkey;Department of Mechanical Engineering, Firat University, 23119 Elazig, Turkey;Department of Mechanical Education, Firat University, 23119 Elazig, Turkey;Department of Electronic and Computer Education, Firat University, 23119 Elazig, Turkey
Venue:
Expert Systems with Applications: An International Journal
Year:
2009

Citing 4
Cited 1

The nature of statistical learning theory

The nature of statistical learning theory
A novel approach for digital radio signal classification: Wavelet packet energy-multiclass support vector machine (WPE-MSVM)

Expert Systems with Applications: An International Journal
Analysis of adaptive-network-based fuzzy inference system (ANFIS) to estimate buoyancy-induced flow field in partially heated triangular enclosures

Expert Systems with Applications: An International Journal
Control of convergence in a computational fluid dynamics simulation using ANFIS

IEEE Transactions on Fuzzy Systems

Housing price forecasting based on genetic algorithm and support vector machine

Expert Systems with Applications: An International Journal

Quantified Score

Hi-index	12.05

Visualization

Abstract

Forecasting of entropy generation of laminar natural convection in a partially cooled square cross-sectional room has been performed using support vector machines (SVM). The two-dimensional room was modeled as floor heating story with a window. Values of temperature and velocities were obtained by solving governing equations of natural convection with finite difference technique and using these values entropy generation was calculated with its definition. Forecasting of entropy generation due to fluid friction irreversibility (FFI) and heat transfer irreversibility (HTI) were made with known values for unknown parameters using SVM. Thus, calculation time was extremely reduced and values were obtained even for non-convergence cases. It was observed that the SVM was a strong method to predict the entropy generation without computational fluid dynamic analysis for all cases.