神经网络在锚杆支护方案优选及变形预测中的应用  

Application of neural network in optimal selection of bolt support patterns and deformation prediction of extraction roadway in fully-mechanized caving face

作　　者：朱川曲[1] 冯涛[1] 施式亮[1] 

机构地区：[1]湖南科技大学能源与安全工程学院,湖南湘潭411201

出　　处：《煤炭学报》2005年第3期322-326,共5页Journal of China Coal Society

基　　金：国家自然科学基金资助项目(50274060);湖南省自然科学基金资助项目(03JJY3077)

摘　　要：根据综放回采巷道的特征,确定其锚杆支护形式优选需要考虑的因素为:围岩强度、煤层强度、巷道埋藏深度、围岩节理裂隙发育程度、采动影响、顶煤厚度、护巷煤柱宽度、巷道断面面积,这些因素和支护形式、支护参数一起对巷道围岩变形产生影响.根据巷道支护方案和围岩变形与其影响因素为非线性关系的特点,建立了综放回采巷道支护方案优选和围岩变形预测的神经网络模型,为支护设计和生产管理提供科学依据.Based on the features of extraction roadway in fully-mechanized caving face,the influential factors such as intensity of surrounding rock, intensity of coal, mining depth, joint and crack of surrounding rock, influence of mining, top coal thickness,width of pillar and area of roadway section were defined in order to select optimally bolt support forms of the roadway. These factors,support forms and support parameters affected the surrounding rock deformation of roadway. On the basis of features of the nonlinear relation between the support patterns of roadway and surrounding rock deformation and their influential factors,the neural network models to select optimally support patterns and forecast surrounding rock deformation of extraction roadway in fully-mechanized caving face were established in order to provide scientific bases for the support design and production management.

关 键 词：神经网络 锚杆支护形式 支护参数 巷道变形 

分 类 号：TD353.6[矿业工程—矿井建设]