A knowledge-based Mamdani fuzzy logic prediction of the motion resistance coefficient in a soil bin facility for clay loam soil

H. Taghavifar, A. Mardani

Research output: Contribution to journalArticle

6 Citations (Scopus)

Abstract

A rule-based Mamdani max–min fuzzy expert system for prediction of coefficient of motion resistance (CMR) is presented. Owing to nonlinear characteristic of soil–wheel interactions, application of fuzzy rule-based models for determination of CMR is instrumental. We were encouraged to apply fuzzy logic approach for the modeling by use of the experience of induced CMR as affected by tire inflation pressure, velocity, and wheel load to be practically applicable with wide range of unknown nonlinear systems. Employment of fuzzy if–then true rules makes it possible to handle ill-defined and nonlinear arrangements and gives higher privilege over conventional methods. Therefore, of aforementioned rules, 27 if–then true rules were incorporated to develop a sophisticated highly intelligent representation based on centroid method for defuzzification stage preceded by Mamdani max–min inference supposition. The model performance was evaluated on the basis of various statistical criteria and also was compared to a conventional model. Mean relative error lower than 10 %, good scattering around line (1:1), and high coefficient of determination (R 2 = 99 %) obtained by the fuzzy logics model are confirmed.
Original languageEnglish
Pages (from-to)293-302
Number of pages10
JournalNeural Computing and Applications
Volume23
Issue number1
Early online date10 Apr 2013
DOIs
Publication statusPublished - Dec 2013
Externally publishedYes

Fingerprint

Bins
Fuzzy logic
Clay
Soils
Fuzzy rules
Tires
Expert systems
Nonlinear systems
Wheels
Scattering

Keywords

  • Fuzzy logic system
  • Coefficient of motion resistance
  • Soil bin
  • Prediction model
  • Artificial intelligence

Cite this

@article{f96c45d63be94bd1a8dfd8d174a4c6d3,
title = "A knowledge-based Mamdani fuzzy logic prediction of the motion resistance coefficient in a soil bin facility for clay loam soil",
abstract = "A rule-based Mamdani max–min fuzzy expert system for prediction of coefficient of motion resistance (CMR) is presented. Owing to nonlinear characteristic of soil–wheel interactions, application of fuzzy rule-based models for determination of CMR is instrumental. We were encouraged to apply fuzzy logic approach for the modeling by use of the experience of induced CMR as affected by tire inflation pressure, velocity, and wheel load to be practically applicable with wide range of unknown nonlinear systems. Employment of fuzzy if–then true rules makes it possible to handle ill-defined and nonlinear arrangements and gives higher privilege over conventional methods. Therefore, of aforementioned rules, 27 if–then true rules were incorporated to develop a sophisticated highly intelligent representation based on centroid method for defuzzification stage preceded by Mamdani max–min inference supposition. The model performance was evaluated on the basis of various statistical criteria and also was compared to a conventional model. Mean relative error lower than 10 {\%}, good scattering around line (1:1), and high coefficient of determination (R 2 = 99 {\%}) obtained by the fuzzy logics model are confirmed.",
keywords = "Fuzzy logic system, Coefficient of motion resistance, Soil bin, Prediction model, Artificial intelligence",
author = "H. Taghavifar and A. Mardani",
year = "2013",
month = "12",
doi = "10.1007/s00521-013-1400-4",
language = "English",
volume = "23",
pages = "293--302",
journal = "Neural Computing and Applications",
issn = "0941-0643",
publisher = "Springer Verlag",
number = "1",

}

TY - JOUR

T1 - A knowledge-based Mamdani fuzzy logic prediction of the motion resistance coefficient in a soil bin facility for clay loam soil

AU - Taghavifar, H.

AU - Mardani, A.

PY - 2013/12

Y1 - 2013/12

N2 - A rule-based Mamdani max–min fuzzy expert system for prediction of coefficient of motion resistance (CMR) is presented. Owing to nonlinear characteristic of soil–wheel interactions, application of fuzzy rule-based models for determination of CMR is instrumental. We were encouraged to apply fuzzy logic approach for the modeling by use of the experience of induced CMR as affected by tire inflation pressure, velocity, and wheel load to be practically applicable with wide range of unknown nonlinear systems. Employment of fuzzy if–then true rules makes it possible to handle ill-defined and nonlinear arrangements and gives higher privilege over conventional methods. Therefore, of aforementioned rules, 27 if–then true rules were incorporated to develop a sophisticated highly intelligent representation based on centroid method for defuzzification stage preceded by Mamdani max–min inference supposition. The model performance was evaluated on the basis of various statistical criteria and also was compared to a conventional model. Mean relative error lower than 10 %, good scattering around line (1:1), and high coefficient of determination (R 2 = 99 %) obtained by the fuzzy logics model are confirmed.

AB - A rule-based Mamdani max–min fuzzy expert system for prediction of coefficient of motion resistance (CMR) is presented. Owing to nonlinear characteristic of soil–wheel interactions, application of fuzzy rule-based models for determination of CMR is instrumental. We were encouraged to apply fuzzy logic approach for the modeling by use of the experience of induced CMR as affected by tire inflation pressure, velocity, and wheel load to be practically applicable with wide range of unknown nonlinear systems. Employment of fuzzy if–then true rules makes it possible to handle ill-defined and nonlinear arrangements and gives higher privilege over conventional methods. Therefore, of aforementioned rules, 27 if–then true rules were incorporated to develop a sophisticated highly intelligent representation based on centroid method for defuzzification stage preceded by Mamdani max–min inference supposition. The model performance was evaluated on the basis of various statistical criteria and also was compared to a conventional model. Mean relative error lower than 10 %, good scattering around line (1:1), and high coefficient of determination (R 2 = 99 %) obtained by the fuzzy logics model are confirmed.

KW - Fuzzy logic system

KW - Coefficient of motion resistance

KW - Soil bin

KW - Prediction model

KW - Artificial intelligence

UR - http://www.scopus.com/inward/record.url?eid=2-s2.0-84888839586&partnerID=MN8TOARS

U2 - 10.1007/s00521-013-1400-4

DO - 10.1007/s00521-013-1400-4

M3 - Article

VL - 23

SP - 293

EP - 302

JO - Neural Computing and Applications

JF - Neural Computing and Applications

SN - 0941-0643

IS - 1

ER -