基于GIS组件与ANN耦合技术的矿井构造复杂程度评价

矿井构造是影响煤矿安全生产的主要因素，对瓦斯突出、矿井突水有着明显的控制作用。矿井构造具有空间性和非线性等特点，使矿井构造复杂程度评价的难度较大，据此利用MapObjects的空间分析功能和非线性人工神经网络（ANN）耦合技术，对矿井构造复杂程度进行评价。根据前人的研究成果，将突水系数、底板倾角变异系数、底板标高变异系数、断层强度指数、断层密度等5项指标引入评价体系，经过空间分析统计和模型识别，建立了矿井构造复杂程度评价模型。在开滦矿区东欢矿8煤层构造复杂程度的评价中，共训练及评价了研究区的226个评价单元．评价结果表明研究区西部及西北部构造复杂，中部及东南部构造相对简单。

Evaluation of Mine Structural Complexity Based on GIS Component and ANN Coupling Technique

Mine structure is the main factor that affected the safety production, and obviously controlled mining conditions such as gas outburst and mine water bursting. Evaluation of mine structural complexity is very difficult because of mine structure has spatial and non-linear features. So the authors have established an evaluation model for mine structural complexity based on coupling technique of non-linear ANN and Mapobjects with spatial analytical function. Based on research findings of forerunners, introduced 5 indices of water bursting coefficient, floor dip angle coefficient of variation, floor elevation coefficient of variation, fault coefficient of strength and fault density introduced into the evaluation system. After spatial analytical statistics and model recognition, established mine structural complexity evaluation model. By this model, the mine structural complexity of No.8 coal seam in 226 units of Donghuantuo mine, Kailuan mining area has been trained and evaluated. The result demonstrated that structures in western and northwestern parts are complex, while in middle and southeastern parts relatively simple.