中文摘要：

为准确评价煤矿立井井筒的安全状态,基于多元统计分析理论,以工程实测数据为训练样本,选取井筒直径、松散冲积层厚度、水位降、卸压槽压缩率、破裂等级、服务年限率、治理方式和冲积层压缩速率8个影响因素作为判别因子,建立了井筒非采动破裂的Fisher判别分析模型。在此基础上,对兴隆庄煤矿副井和东风井的安全状态进行了预测。结果表明,21个训练样本的回代判对率为100%,模型精确性高;判别效果通过了显著性检验,并通过井筒实例对模型的可靠性进行了验证;兴隆庄煤矿副井和风井2017年6月预测状态为安全,井筒破裂的可能性较小。该预测方法简单、可靠,为立井非采动破裂预测提供了一种新的途径。

英文摘要：

In order to accurately evaluate the safety state of coal mine shafts, based on the theory of multivariate statistical analysis, the measured data were used as training samples and eight main influencing factors including shaft diameter, thickness of the alluvial layer, water level drop, compression ratio of the stress-relief slot, degree of shaft fracture, service life rate of shaft, governance mode and alluvial layer compression rate were selected as the diseriminant indexes, and then the Fisher discriminant analysis model of shaft safety state was established. And on this basis, the safety state of the auxiliary shaft and east ventilation shaft of Xinglongzhuang Mine was predicted. The results showed that the correct rate of back substitution of discriminating was 100% and the model accuracy was high. The discriminant effect was tested by significance test, and the reliability of the model was verified by the real shaft examples. The predicted state of auxiliary shaft and east ventilation shaft of Xinglongzhuang Mine in June 2017 was favorable and the shaft was less likely to fail. The prediction method was simple and reliable, which provided a new way for the prediction of non-mining-induced fracture of vertical shafts.