论文部分内容阅读
在实际施工过程中,基坑变形发展与多种因素有关,很难用力学模型进行定量分析,而根据已有的监测数据建立模型预测未来一段时间内的变形发展趋势,进而对当前施工方案进行评价、调整,方法更可靠。首先,分析深基坑施工实际监测数据的发展规律,由经济预向量自回归测VAR模型(vector auto regression,VAR)建立基坑施工变形预测VAR模型,并结合监测数据进行模型精度评价和预测比分析。根据分析结果,模型精度排序为:VAR模型>新陈代谢模型>神经网络模型>GM(1,1)模型,模型预测比排序为:VAR模型>新陈代谢模型>神经网络模型>GM(1,1)模型。
In the actual construction process, the deformation of the foundation pit is related to many factors. It is difficult to quantitatively analyze the deformation of the foundation pit with the mechanical model. The model is established based on the existing monitoring data to predict the development trend of the deformation in a period of time in the future. Evaluation, adjustment, the method is more reliable. First of all, the development rule of actual monitoring data of deep foundation pit construction is analyzed. VAR model of predicting deformation of foundation pit is established by VAR model. Combined with the monitoring data, the model accuracy evaluation and prediction ratio analysis. According to the results of analysis, the order of the model accuracy is: VAR model> metabolic model> neural network model> GM (1,1) model, and the model prediction ratios are: VAR model> metabolic model> neural network model> .