立井爆破效果预测及模糊综合评价  

Vertical Shaft Blasting Effect Prediction and Fuzzy Comprehensive Evaluation

作　　者：王宇涛[1] 刘殿书[1] 梁书锋[1] 李洪超[1] 李明慧[1] 

机构地区：[1]中国矿业大学(北京)力学与建筑工程学院,北京100083

出　　处：《爆破》2014年第3期10-14,27,共6页Blasting

基　　金：教育部博士点基金项目(编号:20100023110001)

摘　　要：为了预测立井爆破效果以优化设计参数,确定以影响立井爆破效果的10个设计参数及"循环进尺、炮孔利用率、平均单耗、超欠挖量、周边孔眼痕率"5个效果参数作为输入、输出层样本,建立基于BP神经网络的立井爆破效果预测模型。结果表明:各参数的预测误差大多控制在5%以内,模型能够较好地达到预测目的;为了对预测结果进行定量化评价,以"炮眼利用率、炸药单耗、超欠挖量、周边孔眼痕率"作为评价指标,建立模糊综合评价模型并应用于白家宫2#副井中,评价结果与现场实际情况一致,达到了定量化评价的目的。将二者结合应用,对于优化立井爆破设计参数具有重要作用。In order to predict vertical shaft blasting effect to optimize the design parameters, ten parameters influencing blasting results and five effect parameters including ＂footage cycle, blast hole utilization, specific charge, quantity of overbreak and underbreak ,rate of smooth blasting hole defect＂ were taken as input layer and output layer samples to establish vertical shaft blasting effect prediction model based on BP neural network. The result indicated that most of the parameters prediction error was controlled within 5% and the prediction model was easy to achieve. In order to evaluate the prediction result quantitatively, the factors such as ＂blast hole utilization, specific charge, quantity of overbreak mad underbreak, rate of smooth blasting hole defect＂ was input as evaluation index to establish fuzzy synthetic evaluation model ,which was applied in 2# auxiliary shaft of BaiJiagong and the evaluation results were consistent with actual situation. The two models played an important role in optimizing vertical shaft blasting design parameters.

关 键 词：立井爆破 BP神经网络 爆破效果预测 模糊综合评价 

分 类 号：TD235.3[矿业工程—矿井建设]