摘要
盾构掘进过程中开挖面压力失衡易导致地表塌陷或隆起的灾难性事故。由于密封舱土压的变化情况与开挖面压力密切相关,因此精确预测及控制密封舱土压是有效预防开挖面压力失衡的关键技术之一。为建立密封舱土压的预测模型,首先从机理上对密封舱土压与掘进参数的关系,特别是刀盘扭矩对土压的影响,做了详细分析,以此为基础确定了密封舱土压模型新的输入输出参数。然后建立了依据密封舱压力传感器数据、基于最小二乘支持向量机(LS-SVM)的土压预测模型。并且基于此预测模型,以密封舱内4点土压预测值与设定值偏差最小为优化指标,采用粒子群算法(PSO)对控制参数进行在线优化,实时控制密封舱土压平衡。最后结合现场施工数据进行了仿真对比分析,验证了模型的有效性和准确性,为准确预测和控制密封舱土压,保证掘进过程安全提供了参考。
During the process of shield excavation, the pressure unbalance of excavation surface may easily cause the collapse or blow-out of ground surface, which may lead to disastrous accidents. Since the change of chamber earth pressure is closely related to the pressure of excavation surface, the accurate prediction and control of chamber earth pressure is one of key technologies for avoiding the pressure unbalance of excavation surface effectively. In order to establish the prediction model of chamber earth pressure, the relationships between earth pressure and excavating parameters, especially the influence of cutter torque on earth pressure, were analyzed in detail. On this basis, the new input and output parameters of earth pressure model were determined. Then, according to the data of pressure sensors on the clapboard, a prediction model of earth pressure was established based on the least squares support vector machine (LS-SVM). And based on the prediction model, by minimizing the deviation between four predicted earth pressures and setting values, the particle swarm optimization (PSO) was adopted to optimize the control parameters, therefore, the online balance control of earth pressure was realized. Finally, in order to verify the validity and accuracy of prediction model, some simulation experiments and comparative analyses were carried out by using on-site construction data. The proposed method provides a technical support for the accurate prediction and control of chamber earth pressure and the safety in excavating process. © 2015, China Coal Society. All right reserved.
出处
《煤炭学报》
EI
CAS
CSCD
北大核心
2015年第12期2979-2986,共8页
Journal of China Coal Society
基金
国家自然科学基金资助项目(61074020)
中央高校基本科研业务费专项资金资助项目(DUT13LAB04)
关键词
盾构
土压平衡
LS-SVM
预测
PSO
Excavation
Forecasting
Particle swarm optimization (PSO)
Retaining walls
Shielding
Support vector machines
Tunneling (excavation)