论文部分内容阅读
盾构施工引起的地表沉降的影响因素众多,给地表沉降的计算带来较大困难,而BP神经网络能较好地建立各个影响因素与地表沉降的非线性关系。为了能得到较准确的地表沉降值,采用Miscrosoft Visual C#和Matlab编制了BP神经网络预测软件。结合盾构施工过程中影响地表沉降的地层几何条件、地层参数以及施工参数,建立了BP神经网络模型,对盾构施工引起的地表沉降进行预测。将该预测模型应用于南昌地铁工程的上软下硬地层中,同时考虑了该地层掌子面泥质粉砂岩所占的比例,并对预测值与实测值的误差进行了对比分析。最终得到的预测结果与实际沉降值较一致,表明该BP神经网络模型可用于类似的工程实践。
Many factors affect the settlement of the earth’s surface caused by shield construction, which brings great difficulties to the calculation of the settlement. However, the BP neural network can well establish the nonlinear relationship between the various factors and the settlement of the earth’s surface. In order to get a more accurate surface subsidence value, using BP neural network prediction software Miscrosoft Visual C # and Matlab. Combined with the geometric conditions of stratum, stratum parameters and construction parameters that affect the settlement of the ground during the construction of the shield, a BP neural network model is established to predict the ground settlement caused by the shield construction. The prediction model was applied to the upper and lower hard formations in the Nanchang subway project. At the same time, the proportion of shaly siltstone in the stratum was taken into account and the errors between predicted and measured values were compared. The final prediction result is consistent with the actual settlement value, which shows that the BP neural network model can be used in similar engineering practice.