基于支持向量机的采空区稳定性分级

针对采空区稳定性分级的影响因素众多且关系复杂的特点，提出采用支持向量机理论对采空区稳定性进行分级。根据分级评价指标选取原则，选取岩体结构、地质构造、岩石的质量指标、地下可见水、地下水体、周边开采的影响、相邻空区的情况、工程布置、跨度、面积、高度、矿柱的尺寸及布置、埋藏深度和采空区的规格14个影响因子，建立了采空区稳定性评价指标体系，引入支持向量机理论，选择有向无环图方式构造多类分类器，得到采空区稳定性分级的支持向量机模型。将该模型用于山东黄金矿业西山矿区的25个采空区进行分级，并与未确知测度方法的分级情况

Classification of goaf stability based on support vector machine

For the reason that many factors with complex relationship influence the classification of goaf stability, the method of support vector machine was proposed. The 14 indexes i.e. , rock structure, geological structure, quality indicators of rocks, visible underground water, groundwater body, effect of peripheral mining, engineering layout, span, area, height, pillar size and layout, buried deep, specification of goaf were selected to establish the index system for evaluation based on the selection principle of grading index. A comprehensive classification model of goaf stability was established by the theory of support vector machine with directed acyclic graph. This model was employed to classify 25 goafs in Xishan mine of Shandong Gold Mining, and the classification states were compared with that by the unascertained measure classification method. The results showed that support vector machine is reasonable , better reflects the practice and can be applied to the actual engineering．