Estimation of heating temperature for fire-damaged concrete structures using adaptive neuro-fuzzy inference system

Hyun Kang, Hae Chang Cho, Seung Ho Choi, Inwook Heo, Heung Youl Kim, Kang Su Kim

Research output: Contribution to journalArticlepeer-review

14 Scopus citations

Abstract

The structural performance of concrete structures subjected to fire is greatly influenced by the heating temperature. Therefore, an accurate estimation of the heating temperature is of vital importance for deriving a reasonable diagnosis and assessment of fire-damaged concrete structures. In current practice, various heating temperature estimation methods are used, however, each of these estimation methods has limitations in accuracy and faces disadvantages that depend on evaluators' empirical judgments in the process of deriving diagnostic results from measured data. Therefore, in this study, a concrete heating test and a non-destructive test were carried out to estimate the heating temperatures of fire-damaged concrete, and a heating temperature estimation method using an adaptive neuro-fuzzy inference system (ANFIS) algorithm was proposed based on the results. A total of 73 datasets were randomly extracted from a total of 87 concrete heating test results and we used them in the data training process of the ANFIS algorithm; the remaining 14 datasets were used for verification. The proposed ANFIS algorithm model provided an accurate estimation of heating temperature.

Original languageEnglish
Article number3964
JournalMaterials
Volume12
Issue number23
DOIs
StatePublished - 1 Dec 2019

Keywords

  • ANFIS
  • Concrete
  • Fire
  • Fuzzy
  • Heating temperature
  • Membership function

Fingerprint

Dive into the research topics of 'Estimation of heating temperature for fire-damaged concrete structures using adaptive neuro-fuzzy inference system'. Together they form a unique fingerprint.

Cite this