Gas Emission Prediction in Underground Coal Mine Using Support Vector Machine
Gas outburst is a complicated nonlinear dynamic phenomenon in coal mines. Gas is always an important factor which influences the safety production of coal mines. It becomes the essential problem of safety production in coal mines at present. Domestic and overseas mining practice shows that occurrences of gas outburst are regional. The disaster occurrence area accounts for only 8％-20％ of whole mine district. The precise prediction of gas emission is very important to coal mines. Gas emission is the synthetic result of stress, gas pressure, and the physical mechanics of coal and so on. It is difficult to determine the gas emission using traditional methods. A novel method, termed Support Vector Machine (SVM), was proposed to predict Gas emission. Based on the statistical learning theory, Support Vector Machine is a new creative learning system, which can find global optimal solutions for problems with high dimensions and non-linearity through small training samples. The model about the relationship between Gas emission and its influence factor is obtained by Support Vector Machine learning from cases history. The result shows this method is feasible, effective and high precision.
gas prediction support vector machine
ZHAO Hongbo RU Zhongliang ZHANG Shike
School of Civil Engineering, Henan Polytechnic University, Jiaozuo 454003 P R China
国际会议
The 2007 International Symposium on Safety Science and Technology(2007采矿科学与安全技术国际学术会议)
河南焦作
英文
1285-1290
2007-04-17（万方平台首次上网日期，不代表论文的发表时间）