中文摘要：

地下矿山开采时，受采矿方法和开采工艺的影响，产生大量采空区，它是影响井下矿山安全生产的主要因素。群空区稳定性的影响因素主要包括4个方面，即：采空区的结构形态参数、群空区的空间分布特征、群空区的应力环境和特征值、采空区的地质环境和矿岩体的物理力学参数，形成了群空区稳定性影响因素的多源信息数据库。为了识别矿山井下空区的稳定性，以CMS空区探测、数值模拟和岩体力学属性为多源信息来源，应用层次分析进行因素权值分析，建立了基于DS证据理论多源信息融合的群空区稳定性分析模型。针对某矿山的工程现状，实现了采空区的4级稳定性分级。研究结果表明：群空区稳定性与结构参数、地质参数和应力环境参数有关，是一典型的多源信息融合系统，运用I）_S证据理论实现了采空区的多源信息融合的稳定性分级，分级结果有利于矿山的安全处置和控制。

英文摘要：

In the process of underground mining, as the effect of mining method and mining technology, massive goals have been formed in underground mines, which is the main factor influencing the goals stability in underground mining. Generally, there are four main factors to affect the goals stability, i. e. structure parameters, space distribution features, stress environment and characteristic value, and geological condition and mechanical parameters of ore and rock mass. There are the sources of the stability identification of goals based on the multiple source information. Based on CMS technology and engineering geology survey, numerical simulation technology, and using the AHP weight analysis, a D-S evidence classification model is built based on the multisource information fusion in the goals stability. Take a mine as example, the mining information and the parameters are collected, and the classifications of goals stability are identified according to a four grade classification method. The result shows that the stability is relative to the structure parameters,geological parameters and the stress environment parameters of goals. It is a typical problem of the multisource information fusion. Using D-S evidence theory, the classification of the goafs stability is identified, and the results are consistent with the goals stability in the engineering recognition. It is an effective approach to indentify the classification of the gaols stability involving many factors.