基于大数据的TBM刀盘切削比能预测研究

Prediction of cutting specific energy of TBM cutter head based on big data

全断面硬岩掘进机(TBM)为目前隧道挖掘工程的通用设备,刀盘为TBM的重要组件,其性能的好坏将直接影响施工进度、施工安全及经济效益。TBM刀盘的切削比能是切削单位体积岩石所消耗的能量,与刀盘破岩效率密切相关,是评价刀盘切削性能的一项重要指标。文中改进了传统的CSM破岩模型,建立了滚刀切削比能计算模型;基于实际工程的施工数据,应用BP神经网络算法,提出了刀盘切削比能预测的方法,并通过工程实例进行了验证。结果表明,预测值与CSM计算值之间的拟合度较高,该比能预测模型可应用于实际工程中。

Currently,Full Face Rock Tunnel Boring Machine(TBM)is a common device for tunnel-excavation projects,and the cutter head is an important component of TBM.Its performance will directly affect the construction progress,construction safety and economic benefits.The cutting specific energy of TBM cutter head is the energy consumed by cutting a unit volume of rock,which is closely related to the cutter head’s rock-breaking efficiency and is an important indicator for performance evaluation.In this article,the traditional CSM rock-breaking model is improved,and the calculation model of hob cutting specific energy is set up.Based on the data on the actual engineering,the BP neural-network algorithm is applied,and the method for predicting the cutting specific energy of the cutter head is put forward;then,this algorithm is verified through a series of engineering examples.The results show that the fitting degree between the predicted value and the calculated value of CSM is high,and the prediction model of cutting specific energy can be applied in practical engineering.