基于ANN导水裂隙带高度预测的过程优化  

Process Optimization of Height Prediction of Water Flowing Fractured Zone Based on ANN

作　　者：赵忠明[1] 刘永良[1] 李祎[2] 董伟[1] 施天威 

机构地区：[1]河南理工大学能源科学与工程学院,河南焦作454003 [2]河南理工大学计算机科学与工程学院,河南焦作454003

出　　处：《矿业安全与环保》2015年第3期47-49,53,共4页Mining Safety & Environmental Protection

基　　金：河南省自然科学基金项目(0411052700);河南省科技攻关计划项目(032421004)

摘　　要：分析了影响导水裂隙带高度的因素,并将其分为主要因素和次要因素,构建了导水裂隙带高度影响因素体系。采用BP神经网络模型,并选取煤层厚度、顶板岩性、煤层倾角、覆岩硬度、工作面斜长、推进速度、岩体碎胀性作为主要因素用于导水裂隙带高度的预测,为了简化预测模型,加快计算速度,在确定的采矿地质条件下可忽略次要因素。预测结果表明,简化的BP神经网络模型能够满足导水裂隙带高度预测的准确度需要,该预测方法可为水体下采煤提供一定的技术指导。In this paper, analysis was made on the factors affecting the height of water flowing fractured zone, which were then divided into the primary and secondary factors, and a system of factors affecting the height of water flowing fractured zone was constructed. BP neural network model was adopted and the thickness of coal seam, the lithological properties of roof rock, the dip angle of coal seam, the hardness of overlying rock, the inclined length of the working face, the advance speed and the bulking deformation of rock mass were chosen as the primary factors for predicting the height of the water flowing fractured zone, Under the determined geological condition, the secondary factors can be ignored in order to simplify the prediction model and accelerate the calculation. The prediction results showed that the simplified BP neural netwook model could meet the prediction accuracy of the height of water flowing fractured zone and this prediction method could provide a certain technical guidance for coal mining under water-bodies.

关 键 词：人工神经网络 导水裂隙带高度 预测 水体下采煤 

分 类 号：TD741[矿业工程—矿井通风与安全]