摘要
应用距离判别分析理论,结合矿井含水层的水化学分析资料,选取6种离子组分的浓度作为突水水源识别的判别因子,建立矿井突水水源识别的距离判别分析模型;以35组采样的水源样品作为学习样本进行训练,建立相应线性判别函数对35组实测数据用回代估计方法逐一进行检验,正确率为97.14%。将建立的模型对待识别的4个样本进行测试,并与实测结果进行比较。此外,利用本文方法对梧桐庄煤矿的突水水源进行了识别。研究结果表明:距离判别分析模型分类性能良好,预测精度高,回代估计的误判率为0.0286。
Based on the principle of Mahalanobis Distance Discriminant Analysis（ DDA）, the distance discriminant a- nalysis model to determine the water-bursting source of mine was established. According to the chemical analysis of the aquifer water in underground shafts, six indexes of water-bursting source determination of mine were taken into account to build a forecast DDA model. Linear discriminant functions were obtained through training of thirty-five sets of in situ data. Each of the thirty-five sets of samples was tested by using resubstitution method according to the DDA model, and the correct rate was equal to 97. 14% after the DDA model was trained. The other four sets of samples were predicted by using this model to test the discriminant ability of DDA model. Moreover, the proposed method was used to deter- mine water source determination of Wutongzhuang Mine water-bursting. The results show that DDA model classification is good, prediction accuracy is high, misjudge rate of resubstitution is 0. 028 6.
出处
《煤炭学报》
EI
CAS
CSCD
北大核心
2010年第2期278-282,共5页
Journal of China Coal Society
基金
"十一五"国家科技支撑计划(2006BAB02A02)
中南大学学位论文创新资助项目(2009ssxt230)
关键词
矿井突水
水源识别
距离判别分析(DDA)
mine water-bursting
headstream recognition
distance discriminant analysis（DDA）
