基于改进的未确知聚类模型的岩爆倾向性预测

发布时间：2018-05-18 23:27

本文选题：安全工程 + 岩爆预测　；参考：《安全与环境学报》2017年01期

【摘要】：在采用未确知聚类评价模型进行多指标分级评价时,常采用置信度识别准则作为待测对象的属性识别,该准则中置信度的取值由人为取定,当置信度取值不同时,得到的分级判定结果往往出现差异,甚至产生完全不同的判定结果。通过距离判别的思想将未确知聚类理论中的置信度识别准则进行改进,并运用到岩爆烈度的分级预测中。根据岩爆发生的主要影响因素,选取岩石单轴抗压强度σ_c、单轴抗拉强度σ_t、最大切应力σ_θ及岩石的弹性变形能指标W_(et)为岩爆主要影响因子。并以σ_c/σ_1、σ_θ/σ_c、W_(ey)为岩爆烈度等级评价因子,建立未确知测度模型,以距离判别改进后的属性识别方法进行分级预测,并与原置信度识别准则得到的判别结果进行分析和比较。为验证改进模型的实用性,以贵州开磷集团马路坪矿区为例,采用改进的未确知聚类模型对其岩爆烈度等级进行预测分析。结果表明,预测结果与实际情况基本吻合,证明采用改进后的未确知测度模型的判别结果不仅消除了由于置信度取值不同造成的判别结果误差,降低了人为主观因素的影响,而且具有较高的判别准确性和可行性。
[Abstract]:When the unascertained clustering evaluation model is used for multi-index classification evaluation, the confidence recognition criterion is often used as the attribute recognition of the object to be tested. In this criterion, the confidence value is determined artificially, when the confidence value is different, The result of grading judgment is often different or even completely different. The confidence recognition criterion in unascertained clustering theory is improved by the thought of distance discrimination and applied to the classification prediction of rock burst intensity. According to the main influencing factors of rock burst, uniaxial compressive strength 蟽 _ c, uniaxial tensile strength 蟽 _ t, maximum shear stress 蟽 _ 胃 and elastic deformation energy index (WSP) of rock are selected as the main influencing factors of rock burst. The unascertained measure model is established with 蟽 _ C / 蟽 _ S _ 1 and 蟽 _ 胃 / 蟽 _ C _ T _ W _ T as the evaluation factor of rock burst intensity. The improved attribute recognition method based on distance discriminating is used to grade prediction, and the result is analyzed and compared with that obtained by the original confidence recognition criterion. In order to verify the practicability of the improved model, the improved unascertained clustering model is used to predict and analyze the rock burst intensity grade in the Maoping mining area of Kaifa Group, Guizhou Province. The results show that the prediction results are in good agreement with the actual situation. It is proved that the improved unascertained measure model not only eliminates the error of the discriminant results caused by different confidence values, but also reduces the influence of human subjective factors. Moreover, it has high accuracy and feasibility.
【作者单位】：中南大学资源与安全工程学院;山东黄金矿业(玲珑)有限公司;
【基金】：国家重点基础研究发展计划(973)项目(2010CB732004) 国家自然科学基金项目(41272304)
【分类号】：TU45

【相似文献】