盾构机土仓压力平衡模型参数辨识方法研究

Model Parameter Identification for Pressure Control of Soil Chamber in Shield Tunneling

为了在盾构机隧道掘进过程中控制土仓压力和地表变形,提出了基于遗传算法的盾构机土仓压力控制模型辨识方法。为了验证控制模型辨识方法的有效性,建立了盾构机土仓压力控制实验台。该实验台由盾构机推进系统、螺旋输送机排土系统、土仓压力与螺旋输送机转速监测系统组成。用于系统模型参数辨识的输入和输出为螺旋输送机的转速和土仓压力。系统辨识的目标函数定义为实验观测的土仓压力与模型计算值残差平方和最小,而模型计算的土仓压力是系统被辨识参数的函数。为了解决参数估计中的局部极小值问题,采用了改进的十进制遗传算法。实验结果表明,与最小二乘法和梯度搜索方法相对比,遗传算法得到的模型参数具有更高的预测精度,预测的土仓压力与观测值吻合的较好。

In order to control the chamber pressure and minimize the ground deformation during shield tunneling, the model-based control strategy and model parameter identification procedure are proposed. To verify the effectiveness of model-based control strategy and model parameter identification procedure, a laboratory testing device for pressure control in shield tunneling has been set up. The laboratory test device consists of thrusting force system, screw conveyor discharging system, excavation advance system, and pressure and screw conveyor rotation speed monitoring system. The system input and output are screw conveyor rotation speed and chamber pressure, respectively. The objective function of system identification is to minimize the difference between the computed chamber pressure which is the function of identified model parameters and the observed chamber pressure. In order to perform minimization of objective function, the genetic algorithm is applied by using MATLAB software. The investigation shows that compared with least square method and the gradient search method, the forecast chamber pressure obtained by genetic algorithm agrees well with observed ones. The effectiveness of control model and model parameter identification procedure is verified.