隧道工程基于实测环境变形的土体参数确定方法

结合有限元数值模拟,以盾构隧道施工过程中环境变形的实测数据为基础,通过引入AMALGAM作为遗传算法,依据Pareto最优理论为评价标准,采用均方根误差进行误差校准,提出了一种基于实测数据得到准确土体参数的方法.针对上海市长江西路越江隧道工程试验段的工程案例,以周围土体侧移和隧道上方地面沉降为目标,以关键土层参数为分析对象进行多目标反分析,并对最优解和实测结果进行验证比较,结果证明多目标反分析精确度优于单目标反分析,并且更接近实际工况.应用基于实测环境变形确定的土体参数分析预测盾构施工对临近桩基础的影响,与测试结果的比较验证了计算方法和参数的可靠性,为隧道工程的环境影响分析提供参考.

Determining Methods for Soil Parameters Based on Field Measurements in Tunnel Engineering

Abstract： In this paper, an efficient multi objective back analysis method of tunnel engineering was proposed by using the AMALGAM as the genetic algorithm, and by using the Pareto optimal theory as the evaluation criterion. The proposed method was applied to a test section of the West Changjiang Road tunnel project. Taking horizontal displacement as one objective, and ground settlement as the other, the soil parameters of the key layers were estimated using this method. The optimal solution was compared with the measured results, which proved that the multi objective back analysis was superior to the single objective back analysis in accuracy. Meanwhile, predictive analysis was taken based on the results of back analysis, and was verified as feasible. This method can provide references for environmental impact analysis of tunnel engineering.