Open AccessMine Fire Prediction Based on WEKA Data Mining
Author(s) -
Xin Wang,
Jian Jun Hao,
Jun Chen,
Weijia Cheng
Publication year - 2019
Publication title -
iop conference series. earth and environmental science
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.179
H-Index - 26
eISSN - 1755-1307
pISSN - 1755-1315
DOI - 10.1088/1755-1315/384/1/012164
Subject(s) - coal mining , confusion matrix , confusion , decision tree , mining engineering , mine safety , c4.5 algorithm , data mining , support vector machine , artificial neural network , engineering , environmental science , coal , computer science , machine learning , waste management , naive bayes classifier , psychology , psychoanalysis
Mine fires are prone to cause catastrophic accidents such as gas accidents and coal dust explosions in gas-exposed coal mines, which seriously threaten the safe production of mining enterprises. In order to effectively reduce the probability of gas explosion in mines, an accurate prediction of possible fire hazards in Linhua Coal Mine is made: collecting data on fire occurrence indicators; using Weka software on the basis of rough set theory, selecting SVM classifier, BP neural network and J48 decision tree to obtain the accuracy of the samples to be tested; analyzing the detailed accuracy, confusion matrix and node error rate, and obtain the optimal algorithm. On this basis, a mine fire predictive control model is proposed to determine and control the hazard grade of mine fire source and effectively reduce the probability of coal mine fire occurrence. The establishment of this model greatly ensures the safe production of gas outburst mine.