A Knowledge-based Electrical Fire Cause Diagnosis System using Fuzzy Reasoning

퍼지추론을 이용한 지식기반 전기화재 원인진단시스템

  • Lee, Jong-Ho (Department of Safety Engineering, Chungbuk National University) ;
  • Kim, Doo-Hyun (Department of Safety Engineering, Chungbuk National University)
  • Published : 2006.06.30

Abstract

This paper presents a knowledge-based electrical fire cause diagnosis system using the fuzzy reasoning. The cause diagnosis of electrical fires may be approached either by studying electric facilities or by investigating cause using precision instruments at the fire site. However, cause diagnosis methods for electrical fires haven't been systematized yet. The system focused on database(DB) construction and cause diagnosis can diagnose the causes of electrical fires easily and efficiently. The cause diagnosis system for the electrical fire was implemented with entity-relational DB systems using Access 2000, one of DB development tools. Visual Basic is used as a DB building tool. The inference to confirm fire causes is conducted on the knowledge-based by combined approach of a case-based and a rule-based reasoning. A case-based cause diagnosis is designed to match the newly occurred fire case with the past fire cases stored in a DB by a kind of pattern recognition. The rule-based cause diagnosis includes intelligent objects having fuzzy attributes and rules, and is used for handling knowledge about cause reasoning. A rule-based using a fuzzy reasoning has been adopted. To infer the results from fire signs, a fuzzy operation of Yager sum was adopted. The reasoning is conducted on the rule-based reasoning that a rule-based DB system built with many rules derived from the existing diagnosis methods and the expertise in fire investigation. The cause diagnosis system proposes the causes obtained from the diagnosis process and showed possibility of electrical fire causes.

Keywords

yager sum;electrical fire;electrical fire cause diagnosis;fuzzy reasoning

References

  1. John D. DeHaan, 'Kirk's Fire Investigation(5th edition)', pp. 305-351, 2002
  2. G.Dvir and G.Langholz, 'Matching Attributes in a Fuzzy Case Based Reasoning', IEEE, pp. 33-36, 1999
  3. 화재보험협회, 전기화재(발생기기별원인), pp. 24-128, 1979
  4. 이종호 김두현, '전기화재 조사를 위한 분류체계 개발', 한국안전학회지, Vol.20, No.3, pp. 53-57, 2005
  5. 이광형, 오길록 공저, 퍼지 이론 및 응용, 홍릉과학출판사, pp. 3.1-4.59, 1991
  6. 최충석 외 5 인 공저, 전기화재공학, 동화기술, 2004
  7. M. Koyuncu, A. Yazici, 'A Fuzzy Knowledge-Based System for Intelligent Retrieval', IEEE Trans. Fuzzy Syst., Vol. 13, No. 3, pp. 317-330, 2005 https://doi.org/10.1109/TFUZZ.2004.839666
  8. 화재조사팀 편저, '현장실무자를 위한 화재원인 조사기법', 인천광역시 소방본부, 2003
  9. 김만건, '전기화재 원인과 예방대책', 손해보험, 2002
  10. 東京消防, 新火災調柤敎本-第3卷 電氣火災編 東京防災指導協會, 2004
  11. Lawrence O. Hall, 'Rule Chaining in Fuzzy Expert Systems', IEEE Trans. Fuzzy Syst., Vol. 9, No. 6, pp. 822-828, 2001 https://doi.org/10.1109/91.971731