Application of light gradient boosting machine in mine water inrush source type online discriminant

• DOI: 10.1504/ijcse.2021.113633
• Publication Date: 2021-03-16T12:53:56Z
• Authors: Yang Yong; Li Jing; Zhang Jing; Liu Yang; Zhao Li; Guo Ruxue
• Keywords: Boosting

Water inrush is a kind of mine geological disaster that threatens mining safety. Type recognition water sources an effective auxiliary method to forecast disaster. Compared with the current hydro-chemistry methodology, it spends large amount time on sample collection. Considering this problem, urgent propose novel discriminate source types online, and further strive create more for evacuation before The paper proposes in-situ discrimination model based light gradient boosting machine (LightGBM). This combined (GB) decision tree (DT) improve network integrated learning ability enhance generalisation. data were collected from sensors such as pH, conductivity, Ca, Na, Mg CO3 components in different bodies LiJiaZui Coal Mine HuaiNan. results illustrate accuracy proposed achieves 99.63% recognise mine. Thus, discriminant timely way online.